Jump to content

Wikipedia:Reference desk/Archives/Computing/2006 November 30

From Wikipedia, the free encyclopedia
Computing desk
< November 29 << Oct | November | Dec >> December 1 >
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.


November 30

[edit]

Web analytics mystery: why is my website's #1 page /nojavascript?

[edit]

I'm using WebTrends On Demand to crunch my website's server logs, and recently, the most accessed page hasn't been our home page as usual, but instead "www.ourdomain.com/nojavascript/" which throws a 404 error. We're talking thousands of requests on that per day, whereas before it was never on the radar screen in the analytics reports. I haven't changed any WebTrends configuration options recently.

I cannot figure out if this is bot activity, an artifact, or what. If anyone knows what might be generating http requests for that URL, I'd greatly appreciate a heads-up.

--WikkiTikkiTavi 03:26, 30 November 2006 (UTC)[reply]

Is it there in the actual logs themselves? -- Consumed Crustacean (talk) 03:30, 30 November 2006 (UTC)[reply]
Thanks for the response. I checked the raw logs (duh - should've thought of that!), and it does appear, but only like 3 times on a day when WebTrends reports hundreds of page views on it. So that does suggest an analysis bug, which I'll take up with support. What I didn't say explicitly is that the "on demand" product I use doesn't actually analyze logs, (I misspoke) but instead is an ASP server-based model that works off a javascript beacon on each page. FWIW, the /nojavascript accesses in the actual log are all from a Firefox browser and for GIF image hits.--WikkiTikkiTavi 23:02, 30 November 2006 (UTC)[reply]
this is really a COMPLETE stab in the dark but possibly its search engine spiders, which may be set to automatically go to a nojavascript page? look to see if theres any files called robot.txt and see if they point to that file as well. Modesty84 22:29, 30 November 2006 (UTC)[reply]
Good idea. There certainly are a bunch of hits on robots.txt, but they look clean. Thanks, though - I appreciate the suggestion! --WikkiTikkiTavi 23:02, 30 November 2006 (UTC)[reply]

Document on ISO/IEC 6592:2000 Guideline

[edit]

Document required giving details on this standard ISO/IEC 6592:2000

Information technology - Guidelines for the documentation of computer-based application systems

You can buy the standard here. –mysid 06:46, 30 November 2006 (UTC)[reply]

Germanium

[edit]

how much germanium is typically in one computer

  • About 0.0016% by weight[1]. How to find stuff like that: some web page somewhere must have a breakdown of materials in a computer. So google for a list of some of the materials that a computer probably contains: germanium gold tin copper computer. Weregerbil 15:52, 30 November 2006 (UTC)[reply]

howto allocate the power of cpu and memory of ram to do job

[edit]

Hi,

howto allocate the power of cpu and memory of ram to do a job?

thank you!

Do you mean electrical power requirements for the CPU ? If not, I don't really understand the question. What "job" needs to be done ? Memory and CPU time are normally allocated automatically by the programs you use, so you don't need to do anything. StuRat 10:26, 30 November 2006 (UTC)[reply]
In some programming languages, you do need to allocate ram (memory), but not the CPU. You can set a priority for a jobs of a job so it will be considered more important (or less important) than other jobs. But, you are not specifically allocating CPU time as other high priority jobs may cut in front of your job. I often set unimportant jobs that take a long time (such as database queries) to extremely low priority so they won't bother me while I do my work. --Kainaw (talk) 14:22, 30 November 2006 (UTC)[reply]

Mounting shared network drive on Linux

[edit]

Is there any way to mount a shared network drive on a Windows server, to a linux box, such that the drive is accessed through something like '/media/zeus/music/'? I don't need specific instructions on how to do so yet, as the installation doesn't yet exist. But I do need general confirmation on can this work on not. 70.88.111.65 13:20, 30 November 2006 (UTC)[reply]

Sure, Samba will do that. -- Finlay McWalter | Talk 13:59, 30 November 2006 (UTC)[reply]
Ah, aren't such things lovely that let you migrate away from those quirky headaches to what you always wanted to have. At the electronics club at my university, the old, fragile (very unstable) Windows server that hosted files and user accounts for the Windows worstations we couldn't get rid of now runs FreeBSD! :-) As a great side effect, home directories and user account are now the same on both Windows and UNIX stations. Sweet. —Bromskloss 20:02, 30 November 2006 (UTC)[reply]

How can I parse this JSON?

[edit]

for example: http://en.wikipedia.org/w/api.php?action=query&prop=revisions&rvprop=content&format=jsonfm&titles=Wikipedia:Sandbox

How can I extract the real content of the page? I can read out until <pages>, but how can I read the sub field of pages? Which name is a number and not definite. I use JavaScript. Yao Ziyuan 14:41, 30 November 2006 (UTC)[reply]

A good understanding of regular expressions would make parsing this file format trivial. This can be had here, and there is a lot of practical advice in the comments here. Droud 01:13, 1 December 2006 (UTC)[reply]

Software for mailing newsletters?

[edit]

What would be a good (preferably free) software or method for sending great amounts of email to people who signed a newsletter? I understand it could be the same software used for spam, but that's not really my goal. It's been a bit tricky to look for this stuff because I keep finding spam software and they're full of crap I don't need or want (tracking, obfuscation, email capturing\generation, etc)

The situation is this: we have a deal with several unrelated websites where hundreds of people will sign up for their newsletters (which we maintain), and we need to send them each an email when needed. It must support simple HTML emails with image attachments, and most important of all, it needs to separate several emails inside categories (since each website will have it's own particular newsletter)

Is there a free software out there that can do this? I guess I could do something like that in PHP, but I'd rather not. Thanks for your help! 200.233.224.44 15:07, 30 November 2006 (UTC)[reply]

MailerMailer provides that service. I have no idea what the cost is. If you attempt it yourself, you should expect to have your email servers blocked rather quickly - requiring you to purchase spammer software to subvert the spam blockers. --Kainaw (talk) 18:15, 30 November 2006 (UTC)[reply]
PhpList is good for this. howcheng {chat} 18:26, 30 November 2006 (UTC)[reply]

apple macbook - please help.

[edit]

Apple's website says-- "Display 13.3-inch (diagonal) glossy widescreen. TFT display with support for millions of colors Supported resolutions: 1280 by 800 (native), 1152 by 720, 1024 by 768, 1024 by 640, 800 by 600, 800 by 500, 720 by 480, and 640 by 480 at 16:10 aspect ratio; 1024 by 768, 800 by 600, and 640 by 480 pixels at 4:3 aspect ratio; 720 by 480 at 3:2 aspect ratio"

I just want to know two things 2) when we keep the screen at 720 by 480 at 3:2 aspect ratio, will a part of screen go black and only the rest of the screen show up? 1) LCDs dont work well in more than 1 resolution. Will macbook screens work well in all above said resolutions or does it work well in only 1280 by 800 which is native?

Thanks.

I'm not sure what happen on MacBook, but on my iMac, each resolution has two options, stretched or not. Yao Ziyuan 17:59, 30 November 2006 (UTC)[reply]
How does it work on such displays when you select a resolution that does not perfectly match the physical pixels on the display? —Bromskloss 20:05, 30 November 2006 (UTC)[reply]
See Native resolution --frothT C 20:07, 30 November 2006 (UTC)[reply]
I'm not sure it added much. Where is the resampling done? On the graphics card? In the driver? —Bromskloss 13:29, 4 December 2006 (UTC)[reply]
I believe it's done on the controller in the monitor itself. If it's a laptop it may be integrated, but it's still separate from the graphics API and drivers. --frothT C 21:47, 6 December 2006 (UTC)[reply]
It will be forced to scale. That it's a mac is irrelevant- it will look bad. But with very low resolutions like 720x480 you probably won't be able to tell --frothT C 20:07, 30 November 2006 (UTC)[reply]

Dirt on a mousepad

[edit]

I've got an optical mouse and every few days it seems these little round spots of dirt show up on the mousepad. Does anyone know what causes this? A co-worker says that they are colonies of microscopic critters that feed on dead skin cells and that the light from the mouse helps them grow, but I'm a bit skeptical. howcheng {chat} 18:24, 30 November 2006 (UTC)[reply]

It's probably just dirt and oil from your hands and the ambient area. I doubt the light of the optical mouse plays any role here—it has too little exposure to any given spot to promote any sort of growing function. --140.247.240.213 18:49, 30 November 2006 (UTC)[reply]
its def just ur standard gook, since ur making the same small circularish movements with your hands all day your just balling up dust and oil and crap Modesty84 22:26, 30 November 2006 (UTC)[reply]

does anybody know a good distrobution of linux for ethical (and non-ethical) hacking?

[edit]

Cheers a lot Andiman

Uh any of them? --frothT C 20:19, 30 November 2006 (UTC)[reply]
List of Linux distributions? --Russoc4 20:56, 30 November 2006 (UTC)[reply]
Slackware is a popular one for hackers. Vespine 21:31, 30 November 2006 (UTC)[reply]
It's not the OS mate, it's how you use it! --Wooty Woot? contribs 03:02, 1 December 2006 (UTC)[reply]
Before my above reply I started writing a big post about the OS being a tool, similar to a paint brush or a guitar, but then I thought maybe the OP is researching for a story? If he wrote the main underground uber hacker protagonist configuring his redhat it might sound a bit sillier then hacking on his slackware box. ;) Vespine 03:33, 1 December 2006 (UTC)[reply]
In that case, Neal Stephenson came up with Finux. --Kjoonlee 16:19, 1 December 2006 (UTC)[reply]
Perhaps one that includes the Metasploit framework. —Bromskloss 13:45, 4 December 2006 (UTC)[reply]
Don't bother, you aren't a hacker. I know this because:
A true hacker wouldn't come to the reference desk to ask such a question
Because a true hacker uses their own OS written in assembly
And true hacker would be using User:Jimbo Wales account to post anyway
:-P Nil Einne 18:23, 5 December 2006 (UTC)[reply]

SQL query

[edit]

I'm trying to figure out how to write an SQL query for a given situaiton. Any help would be appreciated.

I have two tables which we can call info and attribs. The info table has a primary key of the name id, and some text fields (title, author, etc.). The attribs table contains attributes for the records in the info table. It has its own primary key (id), a field which corresponds to the id field of the info table (info_id), and two text fields, one which is the name and the other is the value of a given name/value pair.

So sample data in the two tables might look like this:

info
id | title   | author
-------------------------
01   Hello     Nobody
02   Great     Somebody
03   Cheers    Everybody

attribs
id | info_id | name   | value
----------------------------
01   02        quality  5
02   02        size     10
03   01        quality  6
04   01        size     30
05   03        size     45

So record #01 in the info table has a record referring to something called "Hello" by "Nobody" has a "quality" of 6 and and a "size" of 30). This arrangement allows me to have a functionally unlimited number of different attributes for any given record, which is what I need for this application.

Basically I want to be able to do a SELECT query which will give me information like this:

id | title | quality 
------------------
01   Hello   6
02   Great   5
03   Cheers

Now the hard part of this, that I can see, is that I am not just trying to JOIN based on the name of a field in the attribs table but the value of a field. And notice I want it to also work even if a given record doesn't have a similar correspondence in the other table.

I'm not really sure if this is possible with SQL or not, but it would be great if it was. (If it isn't, well, I'll just have to figure out something else which won't be as quick.) Any thoughts? --140.247.240.213 18:47, 30 November 2006 (UTC)[reply]

select id, title, value as quality from info join attribs on info.id=attribs.info_id where attribs.name='quality' - will give you the id, title, and value for all items, limiting value to those named quality. If you have more than one quality set for an item in info, you'll get two rows. --Kainaw (talk) 19:19, 30 November 2006 (UTC)[reply]
kainaw is the sql prince --22:25, 30 November 2006 (UTC)
This SQL would fail to select the third row in his output example, however an OUTER JOIN would return null values for the missing records. Droud 01:30, 1 December 2006 (UTC)[reply]
(edited) If you're looking for the field names to be dynamically generated, this would require a SQL scripting language such as T-SQL (MSSQL) that allows you to create, manipulate, and return dynamic recordsets. This would be very slow and cumbersome compared to processing the results on the client side. Is there any reason you want/need field names to be created dynamically from data? Droud 01:20, 1 December 2006 (UTC)[reply]

If the tables are small, such that performance is not an issue, I'd just do the bulk of the work in a program module with embedded SQL (which could be as simple as SELECT ALLs from each table). Then, you can easily set column headers to whatever dynamic values you wish. I can provide a code example if you wish. StuRat 06:18, 1 December 2006 (UTC)[reply]

Try this:
SELECT info.id, info.title, attribs.quality
FROM info
LEFT JOIN attribs
ON (info.id = attribs.info_id)
GROUP BY attribs.info_id
HAVING attribs.quality
LEFT JOIN will still select data from the FROM table table even if the LEFT JOIN table has no matches. I've not tested the SQL, but I think this is what you're needing. It would be easier to do it with the code you're querying the data with, but if you're in no rush it will end up being more efficient if you can achieve this with SQL. If the HAVING attribs.quality bit generates an error, try HAVING attribs.quality > 0 (if you don't need values with 0 selected).RevenDS 21:56, 6 December 2006 (UTC)[reply]

Wii Q#2

[edit]

Does Wii come with the Wifi built in or do you have to buy an expansion? Also (tell me if I am wrong) you have to buy a wifi extension to attach to the Wii for the DS to recieve "patches". --Darkest Hour 19:33, 30 November 2006 (UTC)[reply]

The Wii has built-in 802.11g/b wifi. Since the DS also communicates via 802.11b, there's nothing you have to buy to connect the two. The Wii's software doesn't currently support such a connection (companies these days rush products out and patch the remaining features/fixes in later), but once it does it'll work without any extra hardware. -- Consumed Crustacean (talk) 20:25, 30 November 2006 (UTC)[reply]
It's not so much that the Wii software doesn't support it for technical reasons or because it was rushed, it's that there's simply no content which uses the functionality yet. A Pokemon game out next January will be the first title to use the link. Sockatume 21:07, 30 November 2006 (UTC)[reply]
That's not exactly what I meant. Nintendo has said the the Wii will act as a DS Download Station, allowing the download of demos and the like. This has not yet been implemented. -- Consumed Crustacean (talk) 21:12, 30 November 2006 (UTC)[reply]
Ah, right. Well I personally wasn't expecting that at launch (they'd only mentioned it once and it was omitted the official statements on the machine so it was more of a "some day") but I get the point. Sockatume 21:22, 30 November 2006 (UTC)[reply]
But do you have to buy a WI-FI (put proper word here)Transmitter to get the Wii to work with the internet? Or does the Wii come with something to help reduce the cost of the buying Wi-Fi? --Your friend, Darkest Hour 21:30, 30 November 2006 (UTC)[reply]
No, I'm not sure what that means exactly, but there is nothing like that packaged with it. You need to buy your own wireless router, or a USB Wi-Fi access point such as the one Nintendo sells (a router is 100x better though, and wouldn't cost very much more). -- Consumed Crustacean (talk) 21:34, 30 November 2006 (UTC)[reply]
So how much would it cost me for one of those? --Your friend, Darkest Hour 21:42, 30 November 2006 (UTC)[reply]
The routers you can buy for varying prices online (newegg.com has some for as low as $30; I recommend making sure you get one with a good feedback rating, though. Paying less than $40 for a router is a risky proposition, usually). You need to have a broadband connection for it to connect to, of course, but if you have that already they are usually very easy to hook up. --24.147.86.187 01:50, 1 December 2006 (UTC)[reply]
Alternatively you could get the USB to Ethernet adaptor and hook it up to your existing wired router/modem. That won't be around until next year though. Sockatume 22:11, 30 November 2006 (UTC)[reply]

Ken Kutaragi trivia

[edit]

Looking here: http://www.1up.com/do/feature?pager.offset=2&cId=3155393 I find this quote "..The PS3 will instill discipline in our children and adults alike. Everyone will know discipline." Sorry for wasting your time with such a minor question, but can anyone confirm this and please give a link to the original speech. I express admiration for this guy, no sarcasm intended. I'd just like to read more. Thank you.83.100.138.110 20:59, 30 November 2006 (UTC)[reply]

ok problem solved I found it myself - http://www.kotaku.com/gaming/media-criticism/ken-kutaragi-misunderstood-or-just-nuts-111634.php shame.87.102.8.53 16:33, 1 December 2006 (UTC)[reply]

Bittorrent. Halp!

[edit]

My Bittorrent gets stuck on 'Checking for firewall' which means nothing will download... how can I fix it? Please explain this to me in really simple terms and not too many complicated questions, I'm bad with things. Vitriol 21:24, 30 November 2006 (UTC)[reply]

You need to configure your firewall/router to port forward the appropriate port for the BitTorrent client. Splintercellguy 22:12, 30 November 2006 (UTC)[reply]
Too complicated. Vitriol 22:16, 30 November 2006 (UTC)[reply]
first, is your computer connected to a router? easy way to check-> does your computer plug into your cable modem/dsl modem, or into some weird box with flashing lights that plugs into your cable modem/dsl modem and other computers?
second, go into start, my network connections, right click local area network... click firewalls or advanced or something, and check turn off windows firewall. if you are running norton firewalls or somethin else you have to turn them off too.
this is the simplest fix, it can potentially open you up to hackers though, especially if your windows isnt updated. if you do have a router the fix is a lil bit more complicated Modesty84 22:22, 30 November 2006 (UTC)[reply]
It fixed itself. Vitriol 22:48, 30 November 2006 (UTC)[reply]
No, now it's saying "launch_torrent failed UnboundLocalError: local variable 'do_launchdir' referenced before assignment". Vitriol 22:51, 30 November 2006 (UTC)[reply]
I would suggest using Azureus, an excellent BitTorrent client that can configure most routers automatically through UPnP. There's also Port Forward, which is an excellent resource. Droud 01:23, 1 December 2006 (UTC)[reply]
utorrent does basically everything azureus does but it's more responsive and much smaller. No JRE overhead. --frothT C 02:54, 1 December 2006 (UTC)[reply]

If your computer has the shoeld in it a red one in the bottom left corner double click it go to the one where it says not monitored and switch to i ahve a virusscanner/firewall i will monitor myself. It worked for me. Tremello22 15:59, 8 December 2006 (UTC)[reply]