Your files are gone and you need them back.. what a nightmare! How do you get them back?

I cannot count the number of times I have advised my readers to make regular backup copies of their files (in fact, Automate your backup and get some peace of mind was one of the very first articles I wrote). Having copies of your files on some place other than your hard drive (such as a CD/DVD) is the surest way to make sure that some unfortunate mouse click, mechanical failure, or glitch doesn't permanently rob you of important documents, irreplaceable photographs, or costly music downloads. (There. I've nagged you again. Now, will you please run a Backup program and burn some discs?)

But you don't have backup copies of these files... you didn't think you'd ever need them.. but now you do. OK.
As I mentioned in Delete does not erase your data--preventing recovery, if the reason you no longer have access to the file you want is because you emptied your Recycle Bin (Trash) or right-clicked+delete, you may recover your file with an undelete tool.. just so long as too much time has not elapsed.
* It is a very good idea to install an undelete tool on your computer before you need it. If you haven't done so and you don't already have an undelete tool installed, use an online scanner/recovery tool such as Softperfect File Recovery to try to effect your recovery.

Other reasons for missing ("inaccessible") files are varied and numerous.. and I could fill a book book with detailed steps for each variant. (If you've had to reformat your hard drive, for example {try an "unformat" tool}. Data recovery is a large-- and costly --business.) Fortunately, I don't have to write it, it's already been written.

Tip of the day+Today's free link: Visit and bookmark the following Webpage.
PC World Magazine and MSN Tech have teamed up to bring us Kirk Steers' How To Recover Almost Anything, a three page primer on data recovery steps, and advice for which tools to use.
This information is first place to look when you run into a file recovery situation.. having it bookmarked ahead of time is a good idea, because timeliness is a factor in your success or failure.

*** Disclaimer: you may have noticed that I used the words "may" and "can try"; this is because recovering damaged/corrupted, deleted, or "lost" files is as much luck as it is science. Certain things improve your chances of success.. but the only sure method is making copies of your stuff, and keeping them somewhere else.. like in a safe-deposit box.

* Bonus free link of the day: Folks, my friend and fellow Internationally Renown Tech Blogger, Bill Mullins, has posted what I consider to be a very important article regarding the current state of security on the Internet. You, your loved ones, and your friends should (IMHO) take a look at Script Kiddie Bonaza--Do-It-Yourself Phishing Kits

Copyright 2007-8 © Tech Paul. All rights reserved. post to jaanix

Share this post :

Para finalizar, no quiero dejar de mencionar que siempre me quedo con un mal sabor de boca cuando uno de estos tipos nombra las características de esta tecnología para luego terminar recomendando REAISERFS o EXT3, sin antes repetir el chistesito de que al tipo de REISERFS lo metieron en cana... en fin, la razon de este malestar es que XFS muestra una ficha técnica mucho mas interesante que la de sus pares con el detalle que XFS fue planteado con esta estructura desde sus inicios que no varia mucho en la actualidad en su implementación bajo Linux desde su implementación original para SGI - Silicon Graphics Inc. Esto apuntala la teoría de que si las cosas están hechas de forma inteligente desde el principio no se hace necesario el larga una version XFS2.

El enlace de SGI

http://oss.sgi.com/projects/xfs/

Sin miedo a nada la gente de debian dedica una pagina con todo tipo de test sobre varios FS... los resultados de XFS son muy buenos.

http://www.debian-administration.org/articles/388

Este articulo no tiene desperdicio aunque se nota que el autor tiene puesta la camiseta de XFS, incluso da tips de como mejorar el rendimiento

http://everything2.com/node/1479435

]]> http://aurolive.wordpress.com/?p=6 Tue, 17 Jun 2008 16:43:14 +0000 aurolive http://aurolive.wordpress.com/?p=6 Traditional methods for uploading, downloading, or editing remote files predominantly have consisted of using an FTP client. An FTP client works fine for uploading and downloading, but editing remote files is a chore, as you have to download a file, edit it, then upload it again. And privacy with FTP is an issue as well.

More recently, other methods have become available, such as FTP with TLS support to protect authentication credentials, and the use of SFTP and SCP to copy files with encryption. While these solved the privacy issues, the convenience factor remained absent; you still needed to download a file, edit it, then re-upload it.

Using protocols like DAV allowed for the use of sharing files via HTTP or HTTPS, and some FTP clients made editing remote files more transparent — removing the manual steps. Even then, navigating remote file systems was limited to the interface of the FTP or DAV client.

The FUSE project has made editing remote files convenient by allowing local-filesystem-like access to remote files. With plugins like sshfs, which transparently makes an SFTP connection to a remote system seem like nothing more than a local directory, convenience and privacy are both available.

Most Linux distributions today provide FUSE support and also provide the sshfs plugin, so it should be a simple matter to install. Some distributions limit access to FUSE, which can be a wise thing considering how powerful it is. Distributions like Mandriva require that users who are allowed to use FUSE are added to the fuse group, which can be accomplished by executing, as root:

# usermod -G fuse joe

This will add the user joe to the fuse group, granting him permission to use FUSE. When this is done, mounting a remote directory that you normally have access to via ssh or SFTP is a simple matter of executing, as the user:

$ sshfs host:/path ~/localpath

This mounts the directory /path on the remote host host in the directory ~/localpath on the local system. Executing the mount command now, it is plain that the remote directory is mounted:

$ mount

host:/path on /home/joe/localpath type fuse.sshfs (rw,nosuid,nodev,max_read=65536,user=joe)

Navigating to ~/localpath will enter the remote file system, and from here you can use any local tool to manipulate files as you wish. You can copy files using cp, navigate the directory tree using Konqueror, and edit files with vim or OpenOffice.org, etc.

To unmount this directory, use the fusermount command:

$ fusermount -u ~/localpath

The sshfs command can also take a number of extra arguments, such as the -p argument to connect to a port other than the standard port 22. Adding sshfs to /etc/fstab can also allow for persistent remote directory sharing using FUSE and sshfs, instead of other connection methods such as NFS. By default, the FUSE mounts are unique to a single calling user, but with options such as -o allow_root or -o allow_other, you can permit access to the mounts to root and/or other users respectively.

FUSE is a very powerful system, and sshfs is just one plugin that can be used with it. FUSE modules exist to provide read/write access to NTFS partitions, navigating archive files without expanding them first, navigating relational databases as if they were regular file systems, encrypted file systems, and more. More information on FUSE and the various plugins available for it can be found at http://fuse.sourceforge.net/.

The Historical Perspective

The original Mac, back in 1984, didn’t have access permissions. Nor did it have the links or aliases mentioned in the title of this post.

Access permissions are used to keep different users from using (or sometimes even seeing) files. The original Mac assumed that there was only one user, and that you wanted to see and use everything. So there were no access permissions.

Aliases were a new feature in System 7, which was released in 1991. An alias is a special file which records the path to an item (which may be a file, a folder, a disk, or whatever) and (just in case) its actual file specification on disk. When you open an alias, the OS acts as though you were opening the original item. And since both the actual file specification and the path are stored, an alias can figure out what happened when you changed the name of the original or moved it, or open the replacement if the original has been replaced.

Unix-y systems have long had the concept of a “soft link”, which is similar. Apple’s only contribution is the file specification part — soft links only store the path to the original file.

The closest the Classic Mac OS came to access permissions were File Sharing permissions: you could share folders so that different users had read and write access to different folders. These permissions were artificial, and didn’t do anything on the Mac where the files were actually stored.

In Mac OS 9, Apple attempted to add multiple-user capability into the Classic Mac OS. This never caught on, and in any case the default installation did not turn the option on. Before this time, however, Mac OS X (not yet named that) was on the horizon. Apple needed to rework their disk format in any case, in order to remove or expand certain technical limitations which had not been a problem on older drives. Access permissions were added into the new disk format, which was called HFS+. (The previous format was HFS, which had replaced the original MFS in 1985.) Despite the availability of access permissions in Mac OS 9, they were not used — the access scheme in the multiple users feature was again artificial.

Read, Write, and Execute

Unix-y systems use a system in which there are three possible actions for which one may have permission: reading, writing, and execution. These permissions can be applied to anything, and are the levels of permissions supported by Mac OS X in all versions so far.

When applied to a file, they control whether you can do the actions named: read the file, write to the file, and execute it (that is, run it as a program).

When applied to a directory (a disk or a folder), they have slightly different meanings. Read permission lets you see what is in the directory. Write allows you to modify the contents of the directory — add, rename, or remove items. Execute allows you to retrieve items by name if you don’t have read permission.

There are two common shorthands for these permissions levels.

The first is numeric: you treat read permission as 4, write permission as 2, and execute permission as 1. Then you report the sum. So 7 means “read, write, and execute”, 5 means “read and execute (but not write)” and so on.

The second is initials: ‘r’ for read, ‘w’ for write, and ‘x’ for execute. This can be used to add or remove permissions individually without altering the others (as in ‘+r’ or ‘-x’) and it can also be used to indicate current permissions, using a dash to indicate a missing permission: ‘rwx’ for “read, write, and execute”, ‘rw-’ for “read and write (but not execute)”, and so on.

The Traditional Unix Model: Owner, Group, and World

The point of access permissions is to grant access to some people but not others. Mac OS 10.4 and earlier use the traditional model of access privileges, in which every file and directory has three sets of permissions.

The first set of permissions is for the owner. The owner is some particular account on the system, and usually is the same user who created the file.

The second set of permissions is for users who are part of some particular group, from the list of groups accessed by the system. (Mac OS X seldom exposes its group listings through the GUI, but they exist. In Mac OS 10.5, they are slightly less hidden than in previous versions.)

The third set of permissions is for everyone else, sometimes called “everyone” or “world”.

Each of these groups has its own set of read, write, and execute permissions. For this reason, permissions are often given using the numeric scheme described above as a three digit number. The first digit is the permissions of the owner, the second is the permissions of the group, and the third is the “everyone” permissions.

When looking for the access available to a particular user, the OS starts with the “everyone” permissions. First it checks whether the user is the owner of the item. If so, then the permissions for the owner are added. Finally it checks whether the user is in the group for the item. If so, then the permissions for the group are added. It never removes permissions — if “everyone” has read permission, then users in the group will still have read permission even if the group does not.

Mac OS X defaults to permissions of 755 on new folders (full permissions for the owner, read and execute for the group and for everyone) and 644 on new files (read and write for the owner, read only otherwise). Although any particular choice has drawbacks in one circumstance or another, this is a reasonable choice.

(It should be noted that there is one obvious place where this choice can have bad consequences: by default, the permissions on a new user’s home directory include read access, but of the folders inside the home directory, only “Sites” and “Public” have read access. If you add another directory immediately inside your home directory and don’t alter its permissions, other users will be able to see the contents!)

You can also specify a full set of permissions using the letters given above, combined with the additional letters ‘a’, ‘u’, ‘g’, and ‘o’. The letters u and g stand for ‘user’ and ‘group’. The letter o stands for ‘other’ (i.e. the “everyone” permissions), while the letter a stands for ‘all’. The permissions for each letter are set with an equals sign, and each part of the permissions are separated with commas (but no spaces). Thus, for example, the numeric form 755 corresponds to ‘u=rwx,g=rx,o=rx’ (or ‘u=rwx,go=rx’ — you can put the g and o together because they have the same value).

That may seem needlessly cumbersome, but the letter notation allows you to specify relative permissions using + and -. This can be handy when you want to add or remove some particular permission, but leave the rest alone. In that case, putting a + in front of the letters turns the permissions on if they were off, while - turns them off if they were on. So, for example, ‘o+rx’ means “add read and execute permissions to ‘everybody’ if they aren’t already available”, ‘g-w’ means “remove write permission for the group if it is available”. You can even combine this with the = notation to specify permissions relative to the different entities: ‘g=u-w’ translates into “set the group permissions to the user permissions, but without write access if it is available”. As with the equals sign, you can string these relative permissions together: ‘u=rwx,g+rx,o=g-w’ is valid.

Access Control Lists

The traditional model above is entirely workable, but it has a limitation. Suppose that there are two users named A and B, and a group named Z. A is a member of group Z, but B is not. Directory X has owner A and group Z. If B needs read access to directory X, then under the traditional model, there are only three choices:

1. Grant read permission to everyone on directory X
2. Add user B to group Z
3. Create an entirely new group, and add everyone in group Z to it, plus user B, and set the group of directory X to the new group

Option 1 may not be acceptable — there may be reasons to keep directory X hidden from other users. Option 2 may also not be acceptable, because there may be reasons to keep user B out of group Z. And although option 3 works, it means there are now two groups to keep track of — if group Z has periodic changes, the new group will have to mirror those changes, which will be tedious at best, and possibly disastrous if security is important.

This problem can be solved another way, though, by letting go of the owner, group, world concept. An alternative model is what is known as Access Control Lists, or ACLs for short. Although this is an oversimplification, ACLs allow any number of users and groups to be given access permissions, instead of just an owner and a single group.

Mac OS 10.5 and later use a hybrid system when running on HFS+. Every file and directory has a traditional owner and group, but additional users and groups may be added to the list. If a disk with extra permissions is accessed using 10.4 or earlier, the additional users and groups are simply ignored.

Administrators, Root, Sudo, and Authority

Ordinary users can only change the permissions of files and directories for which they are the owner, but nothing else. If the commands to do so are issued, they do not get executed.

There is, however, a way around this: Unix-y systems have a built-in concept called a “superuser”. A superuser is an account which can execute any command at any time. Commonly, there is only one such account, which is named root, and it is the owner of the root directory of the filesystem. Various methods exist for allowing non-superuser accounts to temporarily gain superuser privileges.

(As a security measure, some Unix-y systems — Mac OS X included, at least by default — run in what is called “rootless mode”. That means that there is no actual account named root, so nobody can ever log in and make arbitrary changes at will.)

Instead, there is a special group of users, who can enter their passwords and temporarily run as superusers. These users are known, in Mac OS X, as “administrators”. If you have examined System Preferences, you will have noticed the little checkbox which allows accounts to administer the computer — that’s what it does.

The mechanism for running as a superuser on Mac OS X is, at the lowest level, the “sudo” command; whenever you see an authentication dialog box asking for an administrator’s name and password, you know that the system is thinking “sudo” underneath. The sudo command works on a per-process basis, and only allows a the program to run as a superuser for a short period of time. Usually (though not always), programs which ask for authentication launch a sub-process which then takes over whatever task needed superuser authority — the main program does not get superuser authority.

Changing Permissions (Without the Command Line)

In 10.3 and later, the Finder has a simple (and simplified) interface for changing permissions: just select the file or directory, choose “Get Info”, and look for the section at the bottom of the info window. You’ll need to click the lock icon to authenticate, then you can make changes. (In 10.5 and later, you can even add new users and groups by clicking the little “+” button under the list of users.)

There are just two small problems: first and foremost, the Finder always asks you to authenticate, even if you are the owner of the file. In addition, the Finder does not give you any direct control over execute permission, just read and write. (In point of fact, the Finder doesn’t give you any way at all to add execute permission to files, and bundles it with read permission for directories.)

There are utilities which can give you finer-grained control over access permissions without resorting to the command line, such as Bare Bones Software’s Super Get Info. Most such utilities cost money.

Changing Permissions (With the Command Line)

You can also use the command line to change permissions. The command line costs no money to use (it’s available in the default installation of Mac OS X in the form of Terminal.app) but it can be confusing for a beginner.

I’m not going to go into much detail about the command here, because the command line requires some other information. (I’ll write a tutorial on Terminal.app later.) But if you’ve used the command line already, you will know how to use the information below.

The three commands involved in permissions on the command line are named chmod, chown, and chgroup. You must use sudo to run any of them, and to use sudo you must be logged in as an administrator. The basic form to use is:

sudo chmod [mode in numeric or letter form] [files or directories]
sudo chown [user name or numeric ID] [files or directories]
sudo chgrp [group name or numeric ID] [files or directories]

A Safety Note

An overwhelming majority of the time, the access permissions for files on a Mac are as they should be. It is very rare for things to get seriously out of whack. And it is also very rare for wrong permissions to cause serious problems for the basic operating of the Mac. (And, I might add, if you’re enough of a beginner to need this page, you will almost certainly not know enough to fix any such problem.)

You may have read online that there is a “repair permissions” script from Apple, and that some people recommend running it often, possibly even every day. The argument is that the screen has “repair” in its name, and it always ends up finding some files and directories whose permissions it changes, so therefore the Mac OS must constantly be corrupting its permissions!

This is not exactly true. The script sets permissions to absolute values every time. It does not “think” about permissions, and therefore it regards a directory with extra permissions available to be wrong. (For example: if a directory has read permission, and all its subdirectories have read permission, then it makes no difference whether that directory has execute permission turned on. But doing that sort of analysis would make the script much more complex, so it doesn’t bother.) Unless there is something noticeably wrong with your Mac, and it was working before, you shouldn’t run the repair permissions script.

So when should you consider changing permissions? Normally, you should only change permissions on an item when you want to share it with another user on the same machine, and that user needs to be able to write to it. (You can grant other users read permission automatically by moving things to the “Shared” folder inside the “Users” folder.)

What Have I Left Out?

There are actually three other permissions which can be turned on: “set uid on execute”, “set gid on execute”, and the “sticky bit”. In numeric terms, these are a 4, 2, and 1 before the user permissions. (1755, for example, sets the permissions to 755 and turns on the sticky bit.) In letters, they are ‘s’ and ‘t’ — s is used with the ‘u’ and ‘g’ letters.

These three extra permissions have special meanings, and are fairly rare. I left them out of the discussion because they are not something beginners should ever deal with. But they do exist.

Where To Go Next?

I have not yet finished the tutorial on Terminal.app. When it is ready, I will add a link here. Meanwhile, why not read my post about paths, which are also useful with the command line.

Ini disebabkan, ternyata virus ntuh telah merubah status folder atau file di Flashdisk ntuh menjadi HIDDEN atawa SUPER HIDDEN (menjadi file system). Sehingga jika di buka di Komputer standar otomatis seluruh folder-folder tadi menjadi kasat mata.

Nah, kali ini wongsinting akan memberikan script batch sederhana buat bisa menampilkan kembali seluruh folder dan file yang telah disembunyikan oleh virus tadi. Karena script ini sederhana, maka bagi para expert dan master, silahkan lewat saja, dari pada cuman cekikikan kagak ada juntrungan.. hihihi..

Tahapan

Pertama buka notepad di windows. Selanjutnya isi atau copy script berikut ke notepad tadi:

echo OFF
echo.
echo.
echo.
echo PROSES PENGEMBALIAN FILE DAN FOLDER
echo -----------------------------------
echo by: wongsinting.wordpress.com
echo.
echo (tergantung banyaknya file..)
echo untuk itu...
echo please wait alias tunggu...
attrib -r -s -h *.* /S /D

Selanjutnya Save file ntuh menjadi .BAT misalnya fileku.bat atau munculkan.bat dll. Caranya yakni dengan SAVE AS terus pilih "Save as Type" nya dengan "All Files", baru namain dech filenya sesuka kita, misalnya ya ntuh tadi.. fileku.bat atau munculkan.bat dll.

Selanjutnya taruh file .BAT ntuh yang barusan dibikin ke Flashdisk. Trus langsung eksekusi dech.. dan.. Jreengg.. semua folder atawa file yang sebelumnya disembunyikan sama virus, sekonyong konyong bermunculan kembali.

This image shows the default places you can send your file in Windows XP: compressed file, desktop, mail recipient, My Documents, and 3½" floppy disk. You can remove Send To destinations you never use -- such as the floppy drive if your machine doesn't have one (most newer PCs don't) -- or add destinations you use frequently, in a few simple steps.

Tip of the day: Take control of your menus. This process is much like adding/removing shortcuts from your Start Up folder, which I've discussed in this prior article. First, we need to open the Send To folder, which is a "hidden" folder inside your Documents and Settings folder (to read my article on hiding/unhiding folders, click here). Open your Run dialogue by hitting Windows key+R, or Start >Run, and type in "sendto" (no quotes).
Here you see the Send To shortcuts which appear on your right-click submenu. To remove an item you never use, just drag it to your Recycle bin (I have already deleted the floppy drive). I frequently send files to a folder on another computer on my network, and for purposes of example I am going to demonstrate adding that to my Send To menu -- but this method can be adapted for any location you'd like to send files.
[update 10/1/07: this can method can also include a printer.] Right-click on any blank area in the Send To window and select (click) New, then Shortcut.
Now the Create Shortcut Wizard opens. We need to browse to our new destination so click on the browse button. To choose a destination, click on it and then click OK. To find my folder on the other computer, I 'drilled down' by expanding the plus signs until I could see my folder. Now complete the Wizard by clicking OK, Next, Finish. Now my new shortcut appears in my Send To window.
Now all I have to do to send a file from this machine to my 'storage' machine is right-click on it...
and select "downloads on P3".

Today's free link: Today, some fun: Knight Online is an extremely popular online fantasy game. From site: "Knight Online is the critically acclaimed medieval fantasy MMORPG developed by Mgame and Noah System. Since its introduction in Korea several years ago, Knight Online has thrilled millions of players in over 80 countries. Players choose between El Moradian Humans and Karusian Tuareks, adventuring as rogues, warriors, mages, and priests."

Copyright 2007-8 © Tech Paul. All rights reserved. post to jaanix 

Share this post :

The other day, the proud owner of a new laptop wanted me to replicate his XP set up onto the new Vista machine, and the usual method has been to to use one of the techniques mentioned in the paragraph above. But I didn't. I used an adjunct to Window's built-in Easy Files and Settings Transfer tool, which will be today's free link.
I downloaded this program to both his XP machine and the new Vista machine. Then I connected his XP machine via wireless. Surprise! The XP machine was instantly seen and recognized.

Then I launched the Windows Easy Transfer Companion on the Vista PC and followed the wizard. The two machines established a "transfer" connection and the XP machine transferred its installed programs, and all the files, and all of the owners tweaks and settings (like bookmarks, and custom toolbars). All I did was watch.
This was, by far, the fastest and easiest user state migration I've ever experienced, and truly was painless. This is bad for a PC Tech's bottom line, but great for Vista owners.

Today's free link: When you buy a new PC, you will almost certainly want to transfer all kinds of things from the machine you've been using to the new one. Microsoft has "a companion" for the Easy Files and Settings Transfer tool called the Windows Easy Transfer Companion. It is actually a 'stand-alone'. This tool not only transfers your documents and personalized Settings tweaks, but the programs you have installed. This is a huge time saver.
I did my transfer over the local network, but you can use the other methods of data storage to make the transfer as well-- including USB thumb drives. [Note: while Microsoft still considers this program to be in beta, I have experienced absolutely no hiccups or difficulties at all.]

Copyright 2007-8 © Tech Paul. All rights reserved. post to jaanix

Share this post :

Also check Weblogzz...Ple do comment

Best filesystem in three categories according to http://fsbench.netnation.com/

Best bang for your buck:

• JFS
• XFS

I/O limited applications:

• ReiserFS 4
• XFS
• ReiserFS 3

For CPU limited applications:

• JFS

The author also encourages us to run our own tests to see it for ourselves.

I chose JFS because its the first one the best bang for your buck and I (think I) have a laptop with a limited CPU.

Read the whole thing here: http://fsbench.netnation.com/

UPDATE. Just to be sure I was making the right decision of switching to JFS, I read the other google results and one forum confirms that JFS really uses less RAM than other file systems.

http://www.linuxquestions.org/questions/linux-general-1/what-is-the-best-linux-filesystem-443612/

Facebook: All third-party developers building on Facebook platform are subject to technical and policy restrictions that strictly limit their collection, use and storage of profile information.

When a user adds an application, they agree to the Facebook platform Application Terms of Use, which allows the developer to make requests for access to the information in the user's profile, excluding contact information.

Users are strongly encouraged to report any suspected misuse of information to Facebook. Additionally, users can block individual applications from accessing any of their data, block all applications, or block individual types of information.

We have sophisticated technology and a dedicated team to address inappropriate activity by applications. Access by applications to Facebook user data is strictly regulated and if we find that an application is in violation of our terms and policies, we take appropriate action to bring it into compliance or remove it entirely.

Facebook is committed to user safety and security and, to that end, its Terms of Service for developers explicitly state that applications may not use adware, spyware, or other deceptive techniques. Users should employ the same precautions while downloading software from Facebook applications that they use when downloading software on their desktop.

Click: We appreciate that Facebook policy and terms and conditions may dissuade law-abiding net users. But for those who do not take notice, what does Facebook do to police its policies?

We do not understand how Facebook can protect users' data if it Is being routinely "sucked off" the site to non-Facebook servers?

Facebook: It is not just the policies that dissuade misuse, it is the fact that there is accountability for such misuse. Facebook has an entire Investigations Team that watches the site and removes content and third-party applications that violate Facebook's Terms of Use.

Facebook users also police the site and use the "report" button if they come across violators of our Terms of Use. We are constantly trying to improve efforts to increase accountability so that users can realise the benefits of third party applications and that those that attempt to abuse user trust are filtered out of the system.

Click: Is Facebook reconsidering at present its default security settings for applications at a level beyond day-to-day review?

Facebook: We regularly evaluate and adjust the security settings for third party applications to ensure that Facebook's Terms of Service are not violated.

We take the issue of privacy very seriously and continually strive to provide a trusted environment where users can join networks that are reflective of their real life communities.

Our privacy settings give users a great amount of control on Facebook, and we are constantly looking for ways to employ new methods to ensure user safety and security.

Kesimpulan tidak berhenti disitu saja, aku tertarik dengan fusefs. FUSE (File-system in USErspace) aslinya dari linux namun sekarang package maupun porting nya sudah tersedia untuk sistem operasi lain. Fuse mengimplementasikan suatu mekanisme yang dapat membuat sesuatu (FTP,Database,Encryption,dll) menjadi memungkinkan untuk diiplementasikan fungsinya sebagai file-system.

Contoh sederhana : kita dapat mengakses ftp site menjadi sebuah volume/file-system melalui proses mount dan lebih simpel jika dibandingkan kita membuka program ftp client terlebih dahulu; Atau kita dapat mengakses database MySQL dengan menjadikannya sebuah volume/file-system; Atau mengakses layanan gmail melalui volume/file-system. Kebayangkan ????

Referensi :

http://fuse.sourceforge.net/

NTFS
Advantages:

• Works natively and perfectly in Windows.
• Other operating systems support read support natively.
• Robust and journaled.
• Very high limits on maximum partition size
• Very high limits on maximum file size.
• Generally efficient at large partition sizes.

Disadvantages:

• Requires third party software to allow for write support on Linux and Mac OS X.
• Slower on Linux and Mac OS X because of the use of FUSE.
• Proprietary filesystem patented, and at the whim of, Microsoft.

FAT32
Advantages:

• Works natively and perfectly in all three operating systems.

Disadvantages:

• No journalling.
• Relatively low maximum partition size (2 Terabytes - not attainable yet but will be).
• Inefficient at large partition sizes.
• Maximum file size limited at 4GB.

-- Jono of http://insidethebrackets.blogspot.com/2007/04/whats-best-cross-platform-filesystem.html

The results for the google search 'best cross platform filesystem' would mostly say that you use a FAT32 file system. But I agree with Jono when he chose NTFS over FAT32. Read what he has to say here: http://insidethebrackets.blogspot.com/2007/04/whats-best-cross-platform-filesystem.html

Update: Ubuntu supports writing to NTFS since gutsy gibbon

Ok. There are solutions for managing files. They are called document management systems or project banks or something like that. And they are great, at least most of them. But no matter how good you document handling system is, there will be numerous versions of the files.

Do you ever think about whether this is the latest version of file? If there is a newer version, it's just not in the system yet? I know. So do I.

Then, if the information is created by another program, it's likely that the file format is different. Just for the sake of it. "Our software, our file format!"

Wrong.

This should be "Dear customer, our software tool is so easy, transparent and compatible, that you and your team don't have to worry about the file and file compatibility".

If an architect adds an extention to the building design, the effects should be visible immediately to the other team members. The owner and project manager should see how it affects the cost and schedule, logistics should be able to see the estimated material schedules, procurement should automatically be able to check the availability of required materials when required and the city planning department could comment to the look of new and planned extentions.

But no. Dozens of file revisions, file transfers, manual editing, faxes, meetings, emails etc. etc. The whole thing stops because of files.

That's why I hate files. If I am to work with information delivered to me by someone or something, why do I need to worry about the validity of that information? Why can't I just get the information right away when it's created?

Can't someone just come up with a project system, that instantly stores, transforms and delivers information to right people?

Please.

Now, I want you to rest assured, Dear Reader, and take joy and comfort in, that if and when I run across "bad things" in the world of tech-- I let you know about them. (I think we can all agree that a rip-off qualifies as a "bad thing".)

The writer named names and pointed fingers unabashedly.. and I suspect, had to go back and clean up their language before hitting Send. They were.. um.. passionate in their outrage.
What had caught my eye, however, was the perpertrating  company's name -- it was a (brand) name I think highly of; and frankly, so does rest of the industry.

What had our e-mail writer so irked? The company had ripped them off over a Gigabyte. They had paid for 16 GB's, and actually gotten just under 15. (14.9, to be exact.)
They felt short-changed, flim-flammed, and lied to... and as I may have mentioned, they did not like the feeling.
My letter writer's angst was natural, but misplaced.

When is a Gigabyte not a Gigabyte?
* Ever since the neolithic era of personal computers, way, way, way, back in the Early Days (circa 1984), when dinosaurs still roamed, and "kilo" was king (I'm talking before "mega", and well before "giga") the men who produced hard drives (aka "storage devices") described the size of their products using numbers other humans could understand. That is: 1, 2, 3, 4, 5, etc. of the good-old "Base 10" system [ten fingers and ten toes=20].
So to those guys, a "Giga" is 1,000,000,000... just like it is to you and me.
* Computers, in their relentless and perverse desire to frustrate, confuse, and generally annoy humans, decided not to speak in Base 10. Computers invented their own numbering system and named it "binary".
In this sick, twisted, and baffling numbering system, a one is still "1" (to sucker us, no doubt), but two is "10". And "3" (to us) is "11" (to them). Want a real kick in the head? Four is "100". And if you expect 5 to be "1,000", you're wrong-- it's 101.
A "Gigabyte" to a computer is 1,073,741,824 bytes.

Let's see if I have done a good job and you've followed along: to the guy who is making the storage device, if he makes it big enough to hold 1,000,000,000 bytes of information, he calls it a "Gigabyte storage device" (always has and always will).
When you plug it into your computer, your machine will see 73,741,824 bytes less than what it thinks of as a Gigabyte, and it tells you that you you have ".93 GB's of available space".
In the case of my writer, we multiply the difference by 16... which equals 14.9 GB's of available space.

Let me be clear, the manufacturer did indeed provide a storage device that can hold 16,000,000,000 bytes of data-- 16 GB's. And machines "see" that as 14.9 GB's. So they're both right.. and my writer wasn't a victim of a scam, flim-flam, nor fraud.
It's just Base 10 vs. Base 2.

Yes. I understand.
And despite that, I actually like computers!

[For those of you who would like more of a description of "Gigabyte" than my attempt, click here; and for more on binary, here.]

Today's free link: FreeRip 3, a C/Net Editor's 5-star CD application. Description: FreeRip is an easy to use application that can record digital audio tracks directly from compact discs to PC files. You can save CD audio tracks to CD-quality WAV files or encode them to OGG Vorbis, WMA, MP3 or Flac compressed audio formats. It can also convert/encode audio files from WMA/WAV/MP3/Vorbis/FLAC. FreeRip also lets you adjust track volume and it supports ID3 tagging and CD-Text. Includes MP3 ID3 Tagger.

Copyright 2007-8 © Tech Paul. All rights reserved. post to jaanix

Share this post :

Process Monitor 1.33 è un software della SysInternals (da un po' di tempo acquisita da Microsoft) che mostra, in tempo reale, le attività al file system e le modifiche/accessi dei programmi al Registro di sistema.

Download Process Monitor 1.33 (standalone - non necessita di installazione)

I think it's rather important that you understand, if you don't already, that when you right-click+delete, or drag something into the Recycle Bin, it isn't really erased. Instead, the name/path entry in the file allocation table (the directory used to locate and 'find' files) is altered in a way that tells Windows to no longer display the file and that this (physical) area is now available for future storage. The same thing happens when you take the drastic step of formatting your hard drive -- it isn't "wiped" like taking an eraser to a chalkboard: the Master Boot Table and the file directory are similarly altered, and once that occurs the machine can neither find your files nor your operating system -- the rest of the 1's and 0's are left in place.

It is because of this fact -- that files aren't erased, but their directories and names are altered -- that undelete and unformat utilities can perform their miracles. Instead of ignoring or treating these altered entries as writable space, they (attempt to) deliberately seek them out and rename them back to a recognizable formula, which restores Windows' ability to 'see', find, and display them. So, why couldn't my undelete restore my file? The most likely reason is: because Windows sees the deleted file as usable space, it has written something new in that location -- and now that new 1's and 0's are there, your file really and truly is gone. (The more time that elapses since you deleted the file, the more likely it is that it has been written over.)

[ A quick aside: I look at a lot of systems' hard drives, and I've found that today people tend to be the opposite of me and my space-paranoia, born out of small HDs. It strikes me that they don't "recycle" anything. ]

Tip of the day: Never assume that your data has been erased. In fact, I suggest thinking in an opposite manner: assume that no matter what proactive measures you've taken, your data is on that hard drive. Tell yourself that a knowledgeable person with the right tools, if they get their hands on your hard drive, can read it. (There are some people in this industry who insist that your files aren't really gone until your hard drive has been melted in a blast furnace!) Particularly keep this in mind when the time comes to donate (or otherwise get rid of) your old computer.

If you are security-conscious, and you want to ensure that when you erase something it's really and truly erased (or you are about to donate your old PC) I recommend that, if you don't already have one, you download a free file shredder utility (I will put one as today's free link) and to choose one that offers multiple methods of shredding. What a "shredder" does is it writes new data, and it does it in multiple passes. Typically writing all 1's on one pass, all zeros on the next pass, and then a completely random pattern of 1's and 0's, and so on. It is generally recognized that your shredder should make 6-12 passes.
* If you do this, you can donate your old PC comfortable in the knowledge that only a several thousand-dollar restoration, performed in a sterile lab, might render your personal information readable again. (If you are a corporation, and it's time to throw out your old hard drives, and there's highly sensitive data on those drives...melt them.)

Today's free link: Zilla Data Nuker 2 (Please note: this program is an exception to my rule of always having run and tested the links I suggest. I have not ever needed to download a file shredder as I've always had one bundled into the Utility Suites I have on my machines. However, this application is 5-star rated by Cnet, and I was unable to locate it on any "blacklists". It is the one I would try first.) From Cnet, "This powerful program helps you shred important files & folders so that they cannot be restored & prevent attempts to recover sensitive deleted files from your hard drive by data recovery or forensic software. Shredder allows you to purge, wipe & erase data with methods that far exceed US Department of Defence standards for file deletion (DOD 5220.22). Easily automate the cleaning process with batch files, shortcuts and scheduler. Supports complete folder deletions including subfolders."

*Original post: 6/22/07

Copyright 2007-8 © Tech Paul. All rights reserved. post to jaanix

Share this post :