Wikipedia:Reference desk/Archives/Computing/2009 November 5

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 5

Mpeg vs .avi

What is better quality when ripping off a dvd. .mpeg or .avi? Also, does changing format from .avi to .mpeg assuming you always pick the highest quality affect the quality from avi to mpeg?

Reason why I am asking is because I am attempting to take clips from avi high quality rips and converting the clips to .FLV. when i convert the .avi to clipped .flv the audio and video are out of sync. But, if i convert the .avi to mpeg then create the .flv clip the audio and video issue goes away but I want the highest quality clips. Thanks 142.176.13.22 (talk) 01:39, 5 November 2009 (UTC)

You have to understand first that .mpeg or .avi is just a container. They do not determine the video quality. What determines the video quality is the video codec used inside the file.F (talk) 03:02, 5 November 2009 (UTC)

^^That was quite informative and clarified a lot of things. Thanks for the links. I'm still curious on why an .avi container to .flv does not sync audio yet a conversion from .mpeg(from the original .avi converted to mpeg) to .flV works fine. I figure it has to do with how the audio and video streams behind the container or the converter I am using sucks. Thanks for your information! 142.176.13.22 (talk) 04:06, 5 November 2009 (UTC)

Unfortunately, converting between video formats is, well, still one of those "feels like we are in the 1990s" kinds of technologies. There are a million formats, a million options, and little easy way to distinguish between them other than trial-and-error. In my experience. What are you using to convert to FLV? That might be the first place to start looking when it comes to figuring out what the problem is, and what the options are. --Mr.98 (talk) 14:33, 5 November 2009 (UTC)

Help with javascript please? highlighting links in menus.

Hi, i am trying to make a page on my website a little better with some javascript. I am trying to apply highlighting of links (blue) on mouse hover, and when the user clicks on the link i want it to turn a different (green) color and stay that color. As you will see by my code, i have accomplished this, but if you click a link you will notice the other links are no longer blue on hover. Why is this? is it because i used the script to make the style white, and that over-rides any other styles? Any help in fixing this little bug would be greatly appreciated! Thank you!

My code is here: http://pastebin.com/m20258cb7

Thanks again!

137.81.112.220 (talk) 04:08, 5 November 2009 (UTC)

I don't think you need JavaScript for that. You can just use CSS: [1].--Drknkn (talk) 05:37, 5 November 2009 (UTC)

Using CSS is so much easier to achieve that than Javascript. --antilived^{T | C | G} 06:02, 5 November 2009 (UTC)

They know how to use :hover, as is clear from the code. That is not the issue at hand, read it more carefully. --Mr.98 (talk) 14:46, 5 November 2009 (UTC)

Yes, it seems to involve the menu1, menu2, menu3 turning it white code (if you comment it out, it works more like you want it to). I recommend, just creating a class called menulink_clicked, and then change the class of the clicked one to that (this.className='menulink_clicked'), and the other ones to regular menulink again. That way you aren't actually mucking across with the classes' stylesheet directly. I am not really sure why it is eliminating the :hover instructions you have already put in there (you don't modify them), but it seems to be ignoring them after those lines of code. --Mr.98 (talk) 14:46, 5 November 2009 (UTC)

What's happening is that initially the elements are being colored by the background-color CSS property cascading down to them from A.menulink:hover, but when you assign to that property explicitly you give the element its own individual style, which will always override the cascaded value. Your best choice is to do the class manipulation suggested by Mr. 98. If you really don't want to do that, you must re-set the style properties for each link that wasn't clicked on and handle the hovering/onMouseOver yourself, which will obviously be a pain. --Sean 17:35, 5 November 2009 (UTC)

Okay so heres the deal. In honesty a friend helped me code most of what was in that page, although i have had programming and do understand mostly how it works. I attempted to make the modification suggested (menulink_clicked class) however its not exactly working as i had thought. can someone have a look and correct my errors please? :)

http://pastebin.com/m5a4111d4

137.81.112.176 (talk) 20:54, 5 November 2009 (UTC)

Nevermind! i fixed it! thanks!!

Resolved

137.81.112.176 (talk) 21:13, 5 November 2009 (UTC)

iPhone v AppleMac Address Book

When I try to sync the Address Book with my iPhone, all the contacts are added in duplicate. Why should this happen please? Can I delete the entire contents (only) on my iPhone Contacts, and re-sync, if so how? Any advice would be appreciated please. Thanks in anticipation.--88.110.20.147 (talk) 08:34, 5 November 2009 (UTC)

I had this problem whe I was syncing between my Mac, PC, iPod touch, WinMo smartphone, and Google... somewhere along the line, something in the name or other key information was changed and I was left with duplicate (and at one point triplicate) entries for every contact. The only effective way to fix this is to make a backup of your address book in one location, delete ALL contacts from every other location and disable syncing from the "parent" location you choose. Then get your address book sorted out, enable syncing, and wait a while for your "new" address book to propegate across devices. I also make a backup of my address book fairly frequently so if this happens, I can restore it in a few minutes. 206.131.39.6 (talk) 18:04, 5 November 2009 (UTC)

Thanks for this idea, will give it a go.--88.110.20.147 (talk) 21:06, 7 November 2009 (UTC)

How to get Wikipedia?

I want to use whole of Wikipedia offline. I thought of downloading Wikipedia (as a whole, not as PocketWikipedia). But is is huge. So, is there a way to get it on a DVD Set or CD Set? (I am ready to pay the price, if required). My Thanks in advance. Anirban16chatterjee (talk) 09:19, 5 November 2009 (UTC)

The closest thing I could find is the Wikipedia for schools initiative, which covers about 5000 articles. Given the size of Wikipedia, downloading anything close to the whole encyclopedia would take a large number of DVDs. — QuantumEleven 12:02, 5 November 2009 (UTC)

You might also want to consider how up-to-date the info will be and how many of the articles will be in a vandalised state when the offline copy is made. Unless there is a really compelling reason to buy many DVDs, stick with the online version. Astronaut (talk) 12:32, 5 November 2009 (UTC)

Wasn't there an article in the Wikipedia namespace that talked about the longterm possibilities of a print Wikipedia? I think I remember reading something like that, but I can't seem to find it right now. —Akrabbim^talk 12:46, 5 November 2009 (UTC)

Perhaps you mean Wikipedia:1.0? -- Finlay McWalter • Talk 13:35, 5 November 2009 (UTC)

Yeah, that's the one I read a few years ago. Still inching forward I guess —Akrabbim^talk 13:40, 5 November 2009 (UTC)

It's actually not the sort of thing that I think collaborative editing is good for. It's the sort of thing that a centralized editorial staff could get together in few months. But for a bunch of volunteers... it's gonna be tough slogging, even if the content wasn't itself changing on a regular basis. --Mr.98 (talk) 00:54, 6 November 2009 (UTC)

3452 volumes 18 stacks

The problem with a print Wikipedia is that it would consume literally an entire wall full of shelves (see image above - and note the scroll-bar! There is more off to the right there!). Similarly, with 4.4Gbytes of article text plus an even greater number of talk pages, WP: pages and photos, it's not going to fit on any kind of dismountable media - there isn't an optical disk format or a tape that could hold it all. So you're down to needing some big hard drives or a heck of a lot of flash memory. There has been at least one successful effort to boil down the essentials to something that'll fit on a CD or DVD - but the vast majority of articles are missing and the think is practically devoid of cross-links because of that - also, all of the pictures are reduced to thumbnails that you can't expand so that many important diagrams and maps are illegible. The pocket Wikipedia is a great little gadget - but again, it's missing the pictures, the Talk pages (no RefDesk!), etc. If you're willing to pay the price - buy a Kindle from Amazon. It's not an offline device - but (at least in the USA) it uses free cellular bandwidth - so you can use it anywhere where Amazon supports the service without paying any connection fees. SteveBaker (talk) 13:52, 5 November 2009 (UTC)

You can buy 3TB portable hard drives these days. One of those would hold it all I think.

Wikipedia m:data dumps are available in a bunch of formats including static html files (those are currently somewhat out of date, and don't include the pictures). The download is a few gb, which isn't that bad (you can transfer it overnight) if you have broadband. That's a compressed archive file so it needs quite a bit more hard drive space, but hard drives are quite large these days. I've been wanting to set up a mediawiki instance and a copy of wikipedia on my home computer for a while, but haven't had the time to figure out the real requirements and install all the different software needed. 69.228.171.150 (talk) 18:51, 5 November 2009 (UTC)

According to the data dump site, the June 2008 HTML data dump of the entire English Wikipedia, minus stuff intentionally left out (user data and deleted content) is 14 gigabytes. That's only about five DVDs or one-seventieth of the LaCIE USB hard drive sitting on top of my computer right now, and while it's not small enough to put in my jeans pocket (as a USB stick containing every Commodore 64 game ever commercially released would be), it's certainly small enough to fit in a bag to carry on my travels. The only downside is that even with my high-speed 2 MB/s Internet connection, downloading the entire dump will take almost two hours. JIP | Talk 21:17, 5 November 2009 (UTC)

But that 14 gigs of HTML doesn't include the pictures - only links to them. SteveBaker (talk) 21:55, 5 November 2009 (UTC)

I wonder why that html dump is so big. The XML dumps, even the current ones, are much smaller than that. And enwiki probably has >50% more text now than it did in mid-2008. Still, it's not a download you'd have to do very often, and once you had it, you could share it with your friends by much faster methods (LAN, wifi, hard drives, dvd's). 14gb might barely fit on three 4.7gb dvd's and should easily fit on two 8.5gb dvd's. 69.228.171.150 (talk) 22:16, 5 November 2009 (UTC)

The OP says they want the entire enclopedia, so they probably won't be interested in this, but an alternate solution would be just to get the FA and GA articles. This would drastically shrink the size of the encyclopia and eliminate a lot of the crap. 12.165.250.13 (talk) 16:13, 6 November 2009 (UTC)

Would an AM3 Quad Core AMD computer work OK with Ubuntu?

Would it just be a normal instal, or does it start getting complicated please? 92.29.76.195 (talk) 10:06, 5 November 2009 (UTC)

I don't see any reason why would based solely on the CPU but the other hardware could cause issues Nil Einne (talk) 16:22, 5 November 2009 (UTC)

Normal install, no big deal. 69.228.171.150 (talk) 18:51, 5 November 2009 (UTC)

vista ram

So I have windows Vista and 2 gb of ram. Why is 1 gb of ram always used up even when absolutely no programs are in use and after a fresh install? Is vista really that ram hungry that it needs a full 1gb just to run, when windows 7 seems to be working good on 512mb of ram —Preceding unsigned comment added by 82.43.88.201 (talk) 12:34, 5 November 2009 (UTC)

That's absolutely right. Vista, along with all its accessory processes, normally takes up at least 1 GB of RAM. That is one of the intentions of Win7 - that it would be faster and more efficient than its bloated predecessor. —Akrabbim^talk 13:42, 5 November 2009 (UTC)

See criticism of Windows Vista. ---— Gadget850 (Ed) ^talk 13:46, 5 November 2009 (UTC)

So if I installed Windows 7 on the same computer as my current windows vista, it wouldn't no longer use 1gb of ram it'd only use 512mb? So I'd have more free ram for programs and stuff? —Preceding unsigned comment added by 82.43.88.201 (talk) 15:07, 5 November 2009 (UTC)

All other things being equal, approximately yes. — QuantumEleven 16:25, 5 November 2009 (UTC)

Free advertising?

My friend has a forum website that really needs members. So what is the best free or cheap way to advertise it?Accdude92 (talk to me!) (sign) 14:43, 5 November 2009 (UTC)

Go to other relevant forums, put a link to it in your "sig", and post some quality things that will make others think you are someone they might want to hear more from? The problem is that there are probably a thousand other forums that do similar things -- you need to think about how you are going to let the small number (relatively speaking) of people out there who might be interested in it know that it exists and is worth their time. It is a non-trivial task for new sites in general, especially forums, which require an existing user base before they become seen as worth participating in. --Mr.98 (talk) 15:14, 5 November 2009 (UTC)

I would be careful with something like that. It could easily come across as spamming. At the very least make sure you read the rules first. Better yet get somewhat established in the forum first before you change your sig. Even better, after getting established but before adding it to your sig, ask whether it will be okay in an appropriate place (don't link to the other forum, that will come across as spamming) unless of course it's clear that you shouldn't ask. Nil Einne (talk) 16:21, 5 November 2009 (UTC)

You can probably tell what is acceptable in a sig by looking at other people's sigs. --Tango (talk) 16:29, 5 November 2009 (UTC)

It depends. What's okay for someone who's been with the site for years may not be okay for someone who just joined yesterday or even a week ago. In particular, altho I neglected to mention this earlier my key point was that if you make a bunch of posts which people find unhelpful, offtopic or otherwise unwanted and it appears to them your sole point is to get more posts and you just joined and are advertising a site in your sig they're unlikely to be happy to say the least even if they don't normally care if newbies advertise sites in the signatures. Nil Einne (talk) 17:20, 5 November 2009 (UTC)

Well, I do emphasize that quality is important in such things. A small link with good posts—probably not offensive. Dumb posts, obvious spamming—not going to be effective. --Mr.98 (talk) 17:42, 5 November 2009 (UTC)

I second the "reputation-based" distinction between spamming and informing. "Spam" is a tough nut to define - it's "undesirable" advertisement (whatever that means). So, if you have no reputation and you ask people to check out your cool stuff, then it probably is spam. But if Steve Baker links to his cool toys (Sorry to use you as an example), or even his commercial ventures, we might take it seriously because he's a credible, consistent contributor, and most of the time he's not trumping up his own website. When a reliable, consistent contributor does link to some external site, we take it a little more seriously than if a new guy shows up touting his own projects with every post. Then again, most mass-media advertisements are commercial endorsements from random strangers; I don't know why they work (I suspect they don't, and $385 billion worldwide are wasted each year). Nimur (talk) 18:14, 5 November 2009 (UTC)

Uncompressed image file formats other than .bmp files

What other file formats are there, apart from BMP files, that are uncompressed and can be processed as a raw bit stream? I require to know the exact file structure of these formats.

The results for searching this includes:
[2], [3] —Preceding unsigned comment added by Csanghamitra (talk • contribs) 15:38, 5 November 2009 (UTC)

I believe that, traditionally, uncompressed TIFF is the easiest uncompressed file format for developers to handle. Note that although there is an uncompressed variant of BMP, most BMP files you'll find are compressed (usually, I think, with RLE.) Similarly, some TIFF files are LZW compressed, as noted in the TIFF article. Comet Tuttle (talk) 18:23, 5 November 2009 (UTC)

PS: Template:Compression formats has a whole list of image compression formats that you should go through. Comet Tuttle (talk) 18:27, 5 November 2009 (UTC)

The Netpbm formats are incredibly simple. -- Coneslayer (talk) 18:25, 5 November 2009 (UTC)

Seconded. I love these formats and their associated tools. —Steve Summit (talk) 23:37, 5 November 2009 (UTC)

TIFF files are commonly used by image processing people. They can store uncompressed data. PNG can store with lossless compression (but compressed), depending on whether your software tool supports this feature. Our Comparison of graphics file formats article allows you to sort by compression technique. If you're looking specifically for no compression, your options are narrowed down pretty significantly; lossless compression preserves information but is a bit more work for you as a programmer. "Anything" can be handled as a bitstream, but it sounds like you want to be able to seek to a specific pixel location without decoding any other values - that is a bitmap by definition; and you probably want a .BMP or TIFF container format. Nimur (talk) 18:02, 5 November 2009 (UTC) —Preceding unsigned comment added by Comet Tuttle (talk • contribs)

I believe that Truevision TGA(Targa) can also be handled in a uncompressed way. But, can also use RLE, so you can't depend on any given TGA being uncompressed. APL (talk) 18:45, 5 November 2009 (UTC)

Linux 64 Assembler Error

I keep getting a (Error: suffix or operands invalid for `pop') for the line (pop %ebx) when I try to compile a script for linux. I looked all over the internet and most hints state that this code is incompatible with linux 64 bit assembler. So what exactly do I have to do to compile it correctly in gcc? 70.171.22.194 (talk) 17:00, 5 November 2009 (UTC)asmProgrammer

You could try pop %rbx, but I doubt that will be sufficient. Porting x86 assembly to 64 bits isn't trivial. The calling conventions are different, structure sizes and offsets are different, and you need to understand how a 32-bit quantity is being used to decide whether it should be extended to 64 bits or not. -- BenRG (talk) 17:56, 5 November 2009 (UTC)

If you post example code demonstrating the problem, we'll have some chance of being able to help you. --Sean 21:19, 5 November 2009 (UTC)

Mac/Unix: Open a file from a remote server on command line

Hi all,

There is a server on my network that I have access to. If I want to copy a file to/from there, I can use scp otherserver:/path/to/file.txt file.txt. Is there any way to open up that file for editing on the command line? For instance, on my computer with TextMate installed, I can run mate file.txt. However, I cannot run mate otherserver:/path/to/file.txt. For some reason, the command tries to find a file in /Users/username/otherserver:/path/to/file.txt.

How should refer to a file on a remote server?

Thanks! — Sam 63.138.152.155 (talk) 18:53, 5 November 2009 (UTC)

If the other machine has TextMate installed and TextMate works over X11, then this will suffice:

ssh -X otherserver mate /path/to/file.txt

If not, this might do for you:

ssh otherserver vi /path/to/file.txt

If you're committed to TextMate and not doing this too often, this will work:

scp otherserver:/path/to/file.txt file.txt
mate file.txt
scp file.txt otherserver:/path/to/file.txt

If you are doing this a lot, you probably want to NFS-mount the directory containing the file you're editing:

ssh otherserver
   sudo sh -c 'echo "/path/to your-local-ip-address/255.255.255.0(rw)" >> /etc/exports'
   sudo /etc/init.d/nfsserver reload   # this will vary based on what the other machine is
   exit
sudo mkdir -p /mount/otherserver/path
sudo mount otherserver:/path /mount/otherserver/path
mate /mount/otherserver/path/file.txt

There will probably be some file permissions things that need some fiddling with. Some editors have remote-editing features built in, so maybe that's worth looking at. --Sean 20:48, 5 November 2009 (UTC)

You might find MacFUSE a useful tool. It allows you to open an SSH session and mount the remote server as a "local file" via a file-system over SSH. Then, you can use any tool on your local system to read and write the remote files as if they were local (from command line or otherwise). You can download MacFUSE and its associated tools here. Similar sshfs tools exist if your local system is running Windows or Linux or Unix. All that is required on the server is an SSH service. Nimur (talk) 00:51, 6 November 2009 (UTC)

If you're doing the editing from a Mac, check out TextWrangler and/or BBEdit; both can edit files over an sftp connection (essentially like scp). -- Speaker to Lampposts (talk) 19:17, 9 November 2009 (UTC)

Help with java script (Part 2) Folding items?

You may remember my post from earlier today where i had issues with a menu highlight system using JS. Ive fixed this now thankfully, but i still need to impliment one more thing. I need to make it so that when you click each of my links, there is a function which shows the appropriate table for that specific link. I have tried to do this once before by simply googling the effect i needed, but i only managed to do strange things like unfold a table by clicking ANYWHERE on the links table, which i obviously dont want! Therefore i removed all of the extra code and decided to ask you, the pros!

What is the best way to impliment folding (hiding!) of tables until the user requests a specific table? I do want to hide all tables except the relevant one. I am including my fixed menu code for reference, with a table for an example. Thanks in advance for any help you can give!

http://pastebin.com/m1266827b

137.81.112.176 (talk) 21:22, 5 November 2009 (UTC)

Easy way:

1. First, design the tables with everything maximally expanded. What would it looked like it EVERYTHING were visible?

2. Once you have that, go through and make each of those tables rows that you want to be hidden a class that has the line display: none; on them, which will make them invisible. Make sure each row has a unique id (e.g. <tr id="table_row_1">[etc.]</tr>, and so on).

3. Get a function that will show or hide elements based on their id and put it in the script tags at the top of the page.. Here is a simple one:

function showRow(id) {
	var ctl = document.getElementById(id);
	if(ctl.style.display=='none') {
		ctl.style.display='block';
	} else {
		ctl.style.display='none';
	}
}

4. Lastly, make any link that you want to expand a given row look something like this: <a href="#" onclick="showRow('table_row_5'); return false;">Click to expand table row #5</a>.

That should give you the framework for what you want; you may find you need to modify it a bit depending on the circumstances. --Mr.98 (talk) 00:50, 6 November 2009 (UTC)

If you're planning on getting more into learning Javascript, you should do yourself a favor and get a toolkit like Prototype or jQuery, which makes life enormously easier. It will allow things like, instead of numbering rows manually and mixing your Javascript with your HTML, just giving them a class like "toggleable", then in your setup code doing:

$$('tr.toggleable').each(function(el) { el.observe('click', Element.toggle) })

which says "give every table row with class 'toggleable' an on-click action that toggles its visibility". --Sean 02:09, 6 November 2009 (UTC)

What's a good definition of a .NET assembly?

I'm trying to come up with a good definition of an .NET assembly. I don't particularly care for the definition given in our article. So far, I've come up with the following (which I've gleaned from a variety of sources):

A .NET assembly is a reusable, versionable, and self-describing building block of a .NET application a collection of types and resources that forms a logical unit of functionality. It contains code in the form of Intermediate Language (IL) along with metadata about the assembly. An assembly can also contain resources used by the assembly. There are two types of assemblies: executables (EXEs) and dynamic link libraries (DLLs). An EXE is a program which can be run by the user. A DLL is a library which is dynamically linked to an executable at run-time. When an application is run, the IL code is compiled into machine language by the CLR (Common Language Runtime)'s just-in-time compiler. Although it's technically possible to create assemblies that span multiple files, in most situations an assembly is a single file.

12.165.250.13 (talk) 21:51, 5 November 2009 (UTC)

For what my opinion is worth, so far, I prefer the definition in the article. Your current first sentence is super nondescriptive. May I recommend you take this discussion to the Talk page of that article rather than the Reference Desk, which is more like a place for people to ask factual questions? Comet Tuttle (talk) 22:47, 5 November 2009 (UTC)

The official definition from the MSDN seems to be the best summary of what a .NET assembly is. You should stick to this definition and cite it as a source. Nimur (talk) 00:56, 6 November 2009 (UTC)

I'm sorry, I didn't mean to imply that I wanted to change our article. I'm actually answering a series of (potential) interview questions and "What is an assembly?" is the first question. I didn't like any of the definitions that I stumbled across, so I tried to write my own.

What I don't like about the article's definition is that it says "an assembly is a partially compiled code library". What does "partially compiled" mean? Does that mean that only every other line is compiled? Does it mean that some classes are compiled and others aren't? Or maybe it has something to do with partial classes? Of course, I know what it means but that's only because I already know what an assembly is. Someone who is unfamiliar with JIT compiling could easily be confused by this. The other thing I don't like is that it says an assembly is a "library". Are EXEs really considered libraries?

Anyway, here's my second stab at creating a definition:

A .NET assembly is a computer file generated by compiling source code written in a .NET programming language such as C# or Visual Basic .NET. An assembly contains both code and metadata, and can also contain resources such as text and graphics that are used by the assembly. The compiled code is in the form of Intermediate Language (IL) and cannot be executed directly. Instead, when an application is run, the IL code is compiled into machine language by the Common Language Runtime (CLR)'s just-in-time compiler. The metadata describes the assembly itself including its name, version number, classes, methods, and properties. There are two types of assemblies: executables (EXEs) and dynamic link libraries (DLLs). An EXE is a program which can be run by the user. A DLL is a library which is dynamically linked to an executable at run-time. Although it's technically possible to create assemblies that span multiple files, in most situations an assembly is a single file.

12.165.250.13 (talk) 14:44, 6 November 2009 (UTC)

Upgrading to 7

My laptop is about three years old, but it has a fresh install of Vista (reformatted and reinstalled in September I think). If I upgrade to Win7, what are the downsides to doing an OS upgrade-type install, as opposed to another clean reformatting and fresh install? I'm not looking forward to another big file transfer project. —Akrabbim^talk 22:24, 5 November 2009 (UTC)

Well, an on top upgrade install takes a long time. I have about 20GB of files on my laptop and upgrading took 5 hours. The good point is that I don't have to reinstall all the programs. A clean install is faster and may be more stable, however, you have to backup and restore your files plus reinstall everything. F (talk) 00:27, 6 November 2009 (UTC)

Is there any impact on performance? —Akrabbim^talk 00:57, 6 November 2009 (UTC)

7's way better than Vista. Vista would randomly freeze for a few seconds. 7 does not have that problem. It feels smoother. F (talk) 02:08, 6 November 2009 (UTC)

I know 7 is supposed to be much more efficient than Vista, I am just wondering if there would be a difference between installing over top without reformatting and a clean reinstall. —Akrabbim^talk 03:17, 6 November 2009 (UTC)

I don't use windows but in general I've found it's preferable to do a clean install than an upgrade. If you've been using a system for a while, you've probably made some customizations, that can break assumptions used by the update procedure. Even if the upgrade doesn't fail outright, it can leave things screwed up in subtle ways. And if one reason you're leaning towards upgrading is to avoid copying your user files off your main drive and restoring them, then your backup practices are deficient and you should work on that too, since that drive will eventually crash. Restoring your files for an OS reinstall is a good opportunity to make sure that your backups actually work, and that is something to be welcomed. 69.228.171.150 (talk) 04:28, 6 November 2009 (UTC)

Since the old OS in question is just two months old, it shouldn't matter that much. F (talk) 09:29, 7 November 2009 (UTC)