Computing desk
< August 13	<< Jul \| August \| Sep >>	August 15 >

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.

August 14

Trackballs

Why did the trackball die out? I always found it far easier and more accurate than the touchpad. Sure it was a little bulky, and you had to plug it into the side of your laptop, but I still found it a pleasure to use. Our article seems to account for its decline by "the replacement of mouseballs by direct optical tracking", but from an ergonomic point of view this doesn't seem to tell the whole story. Has any research been done comparing the speed and accuracy of trackball use vs. touchpad use? --Richardrj ^talk ^email 07:10, 14 August 2007 (UTC)[reply]

You can still buy them, they haven't "died out", they just didn't really catch on in the first place. Like dvorak --Lu c id 07:34, 14 August 2007 (UTC)[reply]

Thanks for that, I didn't know you could still get them. Will they work with today's available laptops? --Richardrj ^talk ^email 08:07, 14 August 2007 (UTC)[reply]

They use a standard USB connector and mouse drivers, so yeah, afaik. I don't know if they still make the type that 'clip on to the side' like our article talks about, but they definitely still have trackball mice, even WalMart has a few --Lu c id 09:04, 14 August 2007 (UTC)[reply]

Also, the moving parts mean it is more subject to failure. I prefer the SpaceBall, which used strain gauges to detect how you were attempting to move the ball, even though you didn't actually move it. StuRat 07:39, 14 August 2007 (UTC)[reply]

I don't know about that. The trackballs I see nowadays are optical, so the only real extra moving part is whatever keeps the ball ~~from~~ rolling freely. --Wirbelwindヴィルヴェルヴィント (talk) 16:17, 14 August 2007 (UTC)[reply]

They typically use a roller for that. Just like an old-style ball mouse, they can get hair and other debris stuck around the roller and start to move in a jerky motion. StuRat 17:02, 14 August 2007 (UTC)[reply]

Vista Sound card problem

Hello I recently installed windows vista on my Sony desktop(came with XP) and cannot play audio. I checked the hardware and it appears the system doesn't recognize the sound card, or see it at all. I don't know where to get the drivers and have lost a lot of the papers that originally came with the pc. All I can remember is that Yamaha made it. How do I get sound to start working again

Thanks, Robert

You might need to go to the manufacturer's website, and download drivers for Windows Vista.

Rechargeable CR2032 button batteries

Does anyone know where I can get the batteries and charger ? StuRat 08:55, 14 August 2007 (UTC)[reply]

I buy them at Walmart or KMart next to the watches. They are very cheap, so I've never considered recharging them. -- Kainaw^(what?) 14:42, 14 August 2007 (UTC)[reply]

I want to use them in key finders, where they go dead in a month or two. If I have 10 units and replace each battery once a month, that's 120 batteries a year. If they cost $5 each, that's $600 a year. Now perhaps you see why I want rechargeable batteries. StuRat 16:59, 14 August 2007 (UTC)[reply]

I buy similar batteries in bulk over the internet from dedicated battery supply companies (the same guys who supply small watchmakers). For CR2032 Google finds me someone who'll sell me 20 for £10, so that's about 1 US dollar each. I really don't think they make rechargable batteries that small. -- Finlay McWalter | Talk 22:08, 14 August 2007 (UTC)[reply]

FrontPage html

Is there a free downloadable program to clean up a frontpage html file in order greatly to reduce its size? - Kittybrewster (talk) 11:05, 14 August 2007 (UTC)[reply]

See HTML Tidy for cleaning up bad HTML and HTML Tidier for removing redundant HTML. -- Kainaw^(what?) 14:40, 14 August 2007 (UTC)[reply]

Youtube search

I'm probably revealing my wanton ignorance of Youtube, but does it allow uploaders to specify a copyright license? If so, does Youtube have a search-by-copyright-license feature - that is, results returned are tagged with a certain copyright license, ala flickr? Raul654 14:08, 14 August 2007 (UTC)[reply]

I've uploaded a few videos to YouTube and they have never presented me with such an option. I'm almost certain that it's not possible. Also, when you do upload your content, you basically agree to give YouTube a full license for them to do whatever they want with it (there was some bruhaha about this on a bunch of blogs when someone realised this a few months ago). The actual text is this (from the TOS):

For clarity, you retain all of your ownership rights in your User Submissions. However, by submitting User Submissions to YouTube, you hereby grant YouTube a worldwide, non-exclusive, royalty-free, sublicenseable and transferable license to use, reproduce, distribute, prepare derivative works of, display, and perform the User Submissions in connection with the YouTube Website...

Makes the GFDL look pretty good, doesn't it :) --Oskar 16:46, 14 August 2007 (UTC)[reply]

Knowing the age of a computer

Is it possible to know when a computer was first switched on? Maybe somewhere in the BIOS? Or does it sound rather unlikely? Thanks. --Taraborn 14:51, 14 August 2007 (UTC)[reply]

There may be circumstantial evidence by checking the dates on the files on the harddrive. You can get a guess, but keep in mind that dates on files can be altered. Also, the drive could have been filled with files from a master system and then put in the PC at a much alter time. Also, the harddrive could be reformatted, losing the files with the original date. -- Kainaw^(what?) 15:13, 14 August 2007 (UTC)[reply]

Most modern operating systems keep a boot log. Assuming it never gets cleared (and this is probably an unlikely assumption), it should have a record for the first time the machine was booted with that OS installed. Of course, other than OEM machines (where the HD is pre-imaged) you have to boot the machine to install the OS Raul654 15:17, 14 August 2007 (UTC)[reply]

I'm guessing that you meant not the first time 'ever' but literally when it was first switched on in a day/period (if it has been on a few days). I expect that there will be something to let you know this but a relatively short search on google found nothing. Perhaps looking for an audit trail of the windows account (presuming windows) would help...This (http://www.windowsecurity.com/articles/Auditing-user-accounts.html) seems to suggest such a thing is possible. ny156uk 16:36, 14 August 2007 (UTC)[reply]

There's no definite standard, but a lot of PC manufacturers will run a "burn-in" test in the factory, which is when it was first turned on. It might even be logged in the service log, or stamped on the case. -- JSBillings 18:08, 14 August 2007 (UTC)[reply]

Interesting points. Thanks to all. --Taraborn 23:37, 14 August 2007 (UTC)[reply]

If you look at the BIOS, the BIOS version will sometimes include the date when it was made. Assuming the BIOS has not been upgraded, it gives a good indication of when the computer was made. --Spoon! 13:08, 15 August 2007 (UTC)[reply]

SharePoint vs. wikimedia

My employer recently decided to create an internal wiki for us to share our knowledge base. It was decided to do it with SharePoint, and nobody seems to know who decided it, or why. The documentation available for SharePoint, official and general Web-stuff, isn't too helpful. Can somebody play the devil's advocate and explain the great advantages of SharePoint as a wiki site so I won't feel so screwed? TIA. 206.213.209.31 18:12, 14 August 2007 (UTC)[reply]

I'm going to hazard a guess that your office is using a whole bunch of Microsoft products. That's almost certainly why your managers choose it, they like Microsoft a lot and they are pretty much bound to it. Many businesses look at open source products (like MediaWiki) with suspicion. As for the advantages, you should probably look at the article, but I assume that it integrates very well with Office and other Microsoft products. --Oskar 18:31, 14 August 2007 (UTC)[reply]

A question about DDR RAM

I have two RAMs working together in my computer (one 512 MB RAM and one 128 MB RAM). They have been fitted into adjacent slots. I used a software called CPU-Z which gives all the information about the computer when installed. It has a section called as TIMINGS TABLE. This table is in the following fashion for the 512 MB RAM.

Timings Table

Frequency               166 Mhz      200Mhz
CAS# Latency             2.5           3.0

And various other values have been given which I cannot understand but I am sure would not be relevant to my question. Now the problem is that in the above table there are two frequency values. I have given the exact reproduction of the table and hence there is no heading for the last two columns of the table. For the 128 MB RAM the timings table has 100 and 133 Mhz instead of 166Mhz and 200 Mhz respectively. Is there a frequency mismatch here? If so, is it still good to have the two RAMs working together? What will happen if they have a frequency mismatch and they are made to work together? In the computer properties I still get a total memory of 512+128=640 MB RAM. Please explain. Should I keep them together or not? Thanks for whoever answers this question. I will be very grateful to you.

Email and Applescript

Can anyone please show me the code to send an email with Applescript?

A beginning relational database application?

I'm interested in creating a relational database of spells from the System Reference Document.

Each spell would have to consist of the following fields:

Name — single-line text
School and subschool — pointer to a school-and-maybe-subschool object
(optional, any number of) Descriptor — pointer to a descriptor object
(any number of) Level — pointer to a class object or cleric domain object, plus an unsigned integer
(1 to 5) Components — any subset of "V", "S", "M", "F", "DF" and "M/DF"
Casting time — possible values are "See text", "1 free action", "1 standard action", "# rounds", "# minutes" or "# hours", where # is a positive integer.
Base duration — possible values are "See text", "Instantaneous", "Permanent", "# rounds", "# minutes", "# hours" or "# days" where # is a positive integer.
Extra duration per level — possible values are zero, "# rounds", "# minutes", "# hours" or "# days".
Range — possible values are "See text", "Personal", "Touch", "Close", "Medium", "Long" or "# feet".
Targets — single-line text
Saving throw — possible values are "None", "See text", "Fortitude/Reflex/Will negates/half/partial".
Spell resistance — yes or no
Harmless — yes or no
Object — yes or no
Material component cost — unsigned integer or "See text"
Focus component cost — unsigned integer or "See text"
Description — multi-line text that should ideally be able to include hyperlinks to other spells.

Ideally, the database should know when sorting by duration or casting time that 1 day = 24 hours = 1440 minutes = 14400 rounds, and that "Instantaneous" and "1 free action" should be treated as zero, "1 standard action" as 1/2 round, and "Permanent" as positive infinity.

Would this be possible in OpenOffice.org Base, and how much skill would it require? I have very little experience with relational databases. Neon Merlin 18:35, 14 August 2007 (UTC)[reply]

There is nothing unusual about the type of DB that you are describing, I don't see why any DB application would have a problem describing the model as you outlined it. In fact, most likely any tutorial that you find about programming a basic DB will end up with something close to what you require. Vespine 01:23, 15 August 2007 (UTC)[reply]

Where's the "relational" part of this? It looks like a single table. A relational database requires multiple tables so there can be some field shared between them. For example, you could have a School table with all the schools and save the school's ID in the School field of the Spell table. -- Kainaw^(what?) 02:13, 15 August 2007 (UTC)[reply]

You could do this in OOo Base though in my opinion it is a bad program to start to learn DB programming on — aside from being relatively unused, I have found it to be extraordinarily slow, incomplete, and sometimes buggy. As someone with a LOT of relational database programming skill (MySQL, MS Access, etc.) I find it almost impossible to use productively at this stage (I hope that someday this will not be the case, because it would be great to have an open-source alternative to Access that worked well!!).

As has been pointed out, this isn't really a relational database, because you have only one table. Some of your fields are somewhat complete datatypes, but they could be easily abstracted as numbers (i.e. "None", "See text", etc. could be 0, 1, etc.).

As it is basically a flat database, something like Filemaker Pro would probably be the easiest way to approach it, if you had access to it. Otherwise you could think about programming the data structure in MySQL and the code to handle it in PHP — it is an ambitious first project but not one that would be insurpassably difficult if you got a "learn PHP in 30 days" type book, which usually deal with how to use PHP with MySQL databases. --24.147.86.187 02:25, 15 August 2007 (UTC)[reply]

With any database design you need to take a whole list of things into consideration, one of these many things is what do you intend to do with the database once you've created it? Why this is important is because presumably OpenOffice is going to have its own data format. That means if you ever wanted to try and connect OpenOffice's Database Management System (DMBS) into a program you might write, you may encounter problems. Also, don't forget that what you're putting in your table is just data, with very little or no intelligence. The only intelligence it has is data constraints - it knows when it's dealing with a date or a number or a character string, etc. But for example if you want it to say "infinite" it would depend on how your program/software is ultimately going to interpret "infinite" if your field is for an integer. You could for example enter in '0' as infinite, or '-1'. My point being, is that to add intelligence to your data you're going to have to write software that goes with it -- all depending on what it is that you're trying to do. Finally, just in case you don't know, a database is a very generalised term, but these days refers to a collection of tables (and can include other metadata info like schemas etc) - and what you've provided is only one table. In order to call your database relational you'll need more than one table, and frankly, if you're going to be doing what I think you're going to be doing then you'll definitely need more than one table. You'll need ones like "Players", "Schools", "Subschools", "Gamedata", "Cast spells" etc. One more thing, it's also a good idea to give your table a primary key - that is a field that will always have unique values - you could for example have your Spell's name as its primary key - unless there's any chance you might have to spells with the same name (in Magic: The Gathering for example, cards with spells were sometimes re-released in a new edition with different artwork, and sometimes different wording). That's why you might consider giving each spell a unique ID, such as "SpellID". Okay I'll shut up now. Rfwoolf 13:42, 15 August 2007 (UTC)[reply]

Hard Drive System Conversion?

Hello,

I'm not sure if my Hard Drive is FAT, FAT32 or NTFS, but I want to convert it to NTFS if it's not. Is there any way to check, and if it's FAT, would changing it have any adverse effects? Would there be any file deletion or the like?

Thanks,

Perfect Proposal ^{Speak out loud!} 22:09, 14 August 2007 (UTC)[reply]

Recent versions of Windows have a utility called "convert" for this - see this Microsoft article. But make a backup of your valuable data regardless. -- Finlay McWalter | Talk 22:12, 14 August 2007 (UTC)[reply]

Partition Magic is a commercial utility focused on manipulation of hard disk partition data, including changing file system types without data loss. ERobson 23:07, 18 August 2007 (UTC)[reply]

webcam capture/basic editing.

So I recently happened upon a webcam, I'm thinking of making a few "you tube" style clips, musical in nature. I suppose I need to do two things, one: Is there a good program to capture the webcam into a file, something that lets you chose the bit rate would be nice. Also, I'd then like a SIMPLE program to do very basic video clip editing, things like cutting and splicing, doing simple fades and audio mixing, the audio will probably be recorded not through the web cam. Thanks. Vespine 23:18, 14 August 2007 (UTC)[reply]

for the editing part, the easiest thing to do would be to use windows movie maker, its really easy to use + its free.

Reverse image search

Is there any such thing as a reverse image search? For example, with Google, a subject can be searched not only by websites but also by what images come up related to the query. Is there a search engine that allows you to enter an image, such as a picture of a person or bird, that then attempts to identify the subject by finding similar images? If there is not such a search engine, is this idea even feasible? --Joelmills 23:51, 14 August 2007 (UTC)[reply]

I don't think so, but I've also wanted this feature occasionally.

Feasible as in possible to set up, somewhat, facial recognition software is basically a 'people search engine'. As far as setting it up on a website, there is a 'family tree' gene site thing that will find faces in pictures, and then match them to celebrities, so it's somewhat possible. Having a large amount of people, and managing the server load from all that though would be a logistics nightmare. Something like this might make more sense in 10-20 years, when computing power is significantly cheaper, or as a client-side application, as already exists in places like casinos --Lu c id 01:13, 15 August 2007 (UTC)[reply]

Headway research into this type of application is being made in the field of Artificial neural networks. You also may be interested in Photosynth, not really what you are after, but related to the topic. Vespine 01:18, 15 August 2007 (UTC)[reply]

IBM's Journal of Research and Development [1] wrote about such a system a few years ago (4? 5?). I don't know what's become of their research, though.

Atlant 12:10, 15 August 2007 (UTC)[reply]

In theory a search engine has to first 'crawl' its pages so that it knows what's on them and then feeds this information into its database. When you then do a search, it doesn't go back to those pages, first it goes through its databases and sees what matches. For the type of search that you speak of, the search engine would have to either put a lot of images into its database (think of the huge amount of space - but that's becoming less of a problem) - or, it will have to come up with some kind of data off of the images, like "number of pixels" + "dimensions" + "size" + "contrast" + "number of colours" + "format" etc. etc. I believe that might be possible, but probably would be a significant challenge to create -- all for a very simple search utility that few people would use

I'd also like to comment about the face-matching people were talking about. To my knowledge, these computers take a source image, and then almost literally try match the image with images it already has in its database. So once again the database would have to hold the images (major space). Then, it would specifically trained at faces. For anything else it would probably not be very useful.

So in short, I do believe we have the technology to do something, but it's just not very feasible. Perhaps you should write to google on this one, they might look into developing it. If they do, please come back here and let us all know, so we can call up our lawyers ;) (jokeing). Rfwoolf 13:52, 15 August 2007 (UTC)[reply]

Thanks for the responses - it was all very interesting. --Joelmills 03:41, 16 August 2007 (UTC)[reply]

You might be interested in CBIR and the links contained therein. skip (t / c) 07:41, 16 August 2007 (UTC)[reply]

A reverse image search engine now exits: TinEye

Wikipedia:Reference desk/Archives/Computing/2007 August 14

Contents