Thursday, July 30, 2009
Monday, July 27, 2009
Monday, July 20, 2009
First we have an online service where we store our data. This would hold our e-mail, pictures, documents, videos, music, etc.
We access our data through web applications. Take Google Docs for example. Currently Google hosts both the application and the data. I would like to see that come to an end. In my model I would log into my Google Docs and set up an account. I would give Google Docs the URL of my web storage server and then Google Docs would give me a unique private key. I would simply highlight this key and copy it into my computer's memory. I would then log into my web storage site and select that I want to share my document data. I would paste the private key provided by Google Docs into my storage interface and receive a unique private key from the storage server. I would provide this unique private key back to Google as authorization.
Behind the scenes Google would contact the web storage server, tell them who's account it wants to access, give it both its own private key that is unique to my Google Docs account, and provide the private key that is unique to the storage server. With both of these keys the web storage server can be confident that it is talking to the correct server that I have authorized to access my documents.
With the trust between the two servers setup I could use Google Docs to modify my documents.
I could repeat this process for different online applications. Files would be saved with open standards so that I can be guaranteed that any application on the web can reliably read and write to them.
I could even use two providers for the same service. I could switch back and forth between using Google's picasaweb and Flickr, for example. There is no reason why only one application would need to be tied to a file type.
If down the road Microsoft comes out with a better online application for modifying documents I could go in to my storage service and deny Google access and go through the process of allowing Microsoft. At any point I could ditch Microsoft and go with another service provider.
Likewise, if I want to change my storage provider I should be able to import all my data to a local file and then upload it to a new provider.
Both the web applications and the data should always be synched up to our primary computers. This way if we don't have an internet connection we can still use the application to modify our data, and the next time the computer comes online it will all synch back up. Google's Gears application already allows for application and data synchronization for offline use, and that's exactly how I envision this working.
The benefits to this are:
- No vendor lock-in.
We are free to change providers at any time. It is not like currently where one must choose Microsoft Office because it is the only application suite that can reliably open and close the .doc format which is the document format.
- Access to data from anywhere with an Internet connection.
You no longer have to carry around a thumb drive or go through any such hassle. Your data is available from almost anywhere.
- Enhanced data persistence
Computer crashes will no longer cause you to lose any of your work.
- Enhanced data security.
A lot is made of the security problems of data existing on the cloud. I believe that you are better served with your data in the hands of a team of professionals than in yours. The average PC user is simply the largest security hole that exists today. As long as the average user is in charge of their data, the average user's data is vulnerable. Google is more likely to keep your data secure than you are.
- OS independence.
Whether you are using Windows, Mac, Linux, your cell phone, game console or any other device, all you need is an Internet connection and a browser to get to your content.
- Cheaper computers.
Because most of the hard work is being done by the application server, your computer no longer needs a lot of ram and processing to run your applications. You just need enough resources to run your web browser.
Sunday, July 19, 2009
To boot into Ubuntu we need to first specify the kernel. By default the latest kernel installed is linked to at /boot/vmlinuz. In the /boot folder lies all of the other kernels you have installed as well. At the console type:
kernel /boot/vmlinuzYou can stop there, but if you want a list of all kernels hit the tab key twice and it will list out each kernel. You can start typing any kernel you wish to boot from, but leaving vmlinuz will boot to the newest kernel.
Now all we have to do is tell grub to boot the os.
bootIf you have Windows installed, we can still boot in it as well, however the instructions are different.
First we have to tell grub which hard drive and partition Windows is installed. The first hard drive, as the bios sees it is 0, and the second is 1 and so on. If you only have one hard drive then we know that it is on hard drive 0. Partitions are zero-based as well. If Windows is on the first partition, then it is partition 0, and so on. For our example Windows is installed on the first hard drive, but the second partition. Here is how we enter that.
root (hd0,1)Remember that there is no space after the comma, but there is one after root. If you aren't sure what your options are, simply type everything up to the hd and hit tab. If you only have one hard drive it will auto-complete the "0," If you have more than one it will list the available hard drives. After you have selected a hard drive you can hit the tab key after the comma and it will again either give you 0 if you only have one partition on that hard drive, or it will give you a list of possible partitions to choose from.
The next three commands will bot up Windows:
makeactiveIf you just find yourself back at the grub console, then you probably entered in the wrong hard drive and/or partitions. Keep trying until you find the right location of Windows.
I haven't tested this out, but if have multiple installs of Linux you can use the root and makeactive command to select the hard drive and partition of the Linux you want to boot into, and then use the kernel command to select the kernel on that hard drive and partition. If anyone knows for sure please drop me a line.
First we'll backup the configuration file and then in your editor of choice, probably vi open up /etc/gdm/gdm.conf.
sudo cp /etc/gdm/gdm.conf /etc/gdm/gdm.conf.bakNext find the line that reads:
sudo vi /etc/gdm/gdm.conf
AutomaticLoginEnable=falseIn the first line change "false" to "true" and in the second line append the username you want to auto-login after the equals sign. Here is how mine looked:
AutomaticLoginEnable=trueAfter a reboot you will find that Ubuntu goes straight to the desktop of the user you defined.
There are other settings you can set in here, such as a time login. Most of these are available from the GUI, but feel free to look around and find any other settings you might want to change.
Update: Ubuntu 11.10 Oneric and onwards use LightDM in place of GDM. Instructions for version of Ubuntu running LightDM can be found here.
Saturday, July 18, 2009
- The "Start" menu.
This is very much like KDE's menu. I don't need to spend time finding where stuff is. I simply type in the name of the application I want to use and it pulls it up in the menu for me. If I want to play my new game, "Call of Duty: World at War", I simply type in any of those words and it shows up in the menu. I no longer have to click Start->Programs->Call of Duty->Game, or whatever. I just type War and I have it. The layout of the item on the right is very helpful as well.
I was never a fan of XP's default menu with the Control Panel, My Computer and all the other items in there because it just made for a lot of clutter. Visually, the default Start Menu in XP was very confusing. The new menu is well organized and easy on the eyes for locating what I want.
As I hover over the items on the right, the picture of me changes to a relevant picture of the item I am hovering over. It has a nice fade effect that works out very well.
If you notice to the right of Sticky Notes there is an arrow. If I click on that arrow I get a list of recently opened documents with that application. So if I had Word installed, it would show Word, and I could click on that arrow to get a list of recently opened Word documents. If I click on the word document then it opens it up.
Unfortunately, if you delete a file, it still shows it as a recently opened file, which I think will be a spot of user confusion.
- The Control Panel.
- Boot time and time to go from login to desktop. I haven't timed it yet, but to go from BIOS to login takes about thirty seconds. To go from login to usable desktop is about five seconds. I know once computer manufacturers start loading up all their crap on people's machines before they buy them, this time will go up considerably, but Windows has done their job at making a very clean bootup.
- Taskbar pins.
- Default user. For a default install of Windows XP, the user setup at install time is the administrator. When the computer boots up the default behavior is that anyone can log in as that administrator without a password. This is horrible security, and I think it is what causes so many people to get infected so very easily. That was Microsoft trying to be user-friendly at the cost of security, which caused more user experience issues than it solved. Now the default account has a password setup at install time, and nobody can log in as that user without the password. I hope this results in more people setting up multiple users on the machine.
I have always like how OSX is so easy to configure. Everything is laid out in a very logical fashion that makes finding the setting you want to modify easy to find. Windows and Ubuntu have always made simple tasks very confusing to find. I think that this new Control Panel simplifies everything. I have found that anything I want to change was easy to do by drilling down through each logical item in the control panel. Again, Microsoft has hit the nail on the head with this one.
There are two improvement here. In the location bar at the top, you can simply go back down the folder tree by clicking on any of the previous folders listed. Ubuntu has this, but I always turn it off. But Windows is doing domething different here. If I click in the location bar I can still type a location manually. This probably doesn't help the majority of people out there. However, I am constantly typing full locations in the location bar, because for me I can do that faster than I can click on a bunch of folders. In Ubuntu your option is the click view or the type view. I like the trade-off between the two views that are present in 7.
The next improvement is the search functionality. In the top-right I can start typing a search, and it will give me the results for the folder, and subfolders that I am currently in. This makes finding the exact file I am looking for very easy.
By default IE, Explorer and Media Player are pinned down in the bottom-left, though they can be removed and other programs added. If I have multiple folders opened, I can click on the pin and it gives me a preview of each open folder. If I hover over one of those previews, the desktop goes black and the full image of the folder is shown to me. If I hover over another folder then that folder is shown to me. If I click on one, then it becomes the active window. This makes finding the right folder very easy. The same is true with IE and having multiple web-pages open.
When creating a new user it defaults to setting those users as "user" and not "administrator" which should help things.
I know this was added in Vista, but I'm still thrilled about it. Updating Windows is no longer tied to Internet Explorer. Updates are done through the control panel. I still have problems with how updates are handled beyond that, which I list below.
- UAC. UAC is still dumb. It is bad security practice. All it will do is teach people to click Yes or Accept more because they will be so tired of trying to read every dialog box. This is Microsoft again refusing to adept sane security because they don't want to compromise usability, even though it is going to get users infected and cause more problems than it solves. One example of how stupid this is, I was trying to install a program and it kept giving me errors all over the place during the install. Once I cancelled I got a box that reading something to the effect of, "It appears you tried to install a program that needs Administrative right. Would you like to rerun this program with Administrative right?" Getting errors all over the place, and then given the option to do it right is not good. Windows should have recognized the need for administrative rights the first time it tried to access a restricted area and prompted me then. And it should have asked for an admin password to ensure that the user had legitimate rights to do that.
- FTP. FTP access from within Explorer is still broken. I have been using 7 while working on some class projects and I couldn't get to me FTP site in Explorer. I could from the command line version of FTP, which is what I ended up doing.
The inability to SSH across computers is another security issue. SSH is kind of like FTP (but certainly not exactly) but it is encrypted, so security is maintained. In the Unix world (which includes OSX) SSH is very powerful, and makes many tasks over the network very easy and secure.
- Updates. Updates still make you reboot constantly. It has taken me as many as four reboots to get all of the updates I needed installed. Why can't it install all the updates the first time, and only make me reboot for very low-level system updates like kernel updates? When I update Ubuntu I do it one time, and I very rarely need to reboot. In fact, I almost never need to reboot.
If you don't reboot in Windows after an update then it will keep nagging you. I griped about there here. I don't understand why Microsoft has such a hard time getting it right. People have been complaining about this since Windows 95, and back then Apple had already made this a non-issue for their OS.
Microsoft should also find a way that users can point to 3rd party update sites, so that users can keep their other software up-to-date using the built-in updating system. So if I have product A, I can point the OS to their updating site and I can get updates to product A with my regular Windows updates.
So there you have it. I still find Ubuntu to be a vastly superior operating system in so many aspects. There are still many questions about Windows 7, such as how well it will run on MIDs, Netbooks, and other lower-end hardware, which we will see pan out after it releases this October.
Wednesday, July 15, 2009
Tuesday, July 14, 2009
Thursday, July 9, 2009
So for a couple weeks I've been brewing in my head a blog post about how what Linux really needs is for Google to come in and work it's magic, but that will never happen because it doesn't fit into Google's business model. So I've been thinking about this for a while and was hoping to post something real soon when just yesterday Google announced they are gearing up to do what I've been working on posting they will never do, which is build their own Linux-based operating system.
So today, we're announcing a new project that's a natural extension of Google Chrome — the Google Chrome Operating System. It's our attempt to re-think what operating systems should be.
That's right, Google is going to have it's own OS. For now it is just aimed at netbooks, and maybe that is as far as it will go. However, the increased influx of Linux development Google will be putting behind Linux should increase as it works to ensure that the new OS works as great as every other Google product.
I am curious to see how this all plays out. I have read some say that this is just another Linux distribution, but that comes from a short-sighted view of what Linux is. It is not an operating system, it is a kernel which an operating system can be built on top of. Nobody pays any attention when they use their TIVO, gPhone, router, or other Linux powered device that they are using Linux. They are simply paying attention to the fact that they are using their device, which is exactly how a device should be.
Google is probably the only company that has the money, power, clout, commitment, and proper understanding of how to utilize open source effectively to make this work. Canonical has made a really good stab at it, but they simply are no Google.
Thursday, July 2, 2009
As a child I spent a lot of time at my grandparents house in Pensacola. They liked to travel, and when they traveled camping was more-or-less their only means of resting. My grandfather owned a 70s model Volkswagen Bus with all the camping accessories and we all loved it.
I had always wanted one, but they have that dirty hippie history behind them, and people already seem to project that idea on me as it is.
So while looking around on the Internet at a vehicle to replace my Oldsmobile Silhouette I came across a 1980 VW Vanagon and I ended up purchasing it.
While I get plenty of weird stares and occasional hoops from other motorists and pedestrian, I have found that by owning a bus I am also now part of a larger community. This community includes previous bus owners, bus admirers, and what I have termed member of the "free riding culture" which includes bikers and others who ride the open road as sport.
I find people approaching me while I pump gas, as I enter and leave stores, and at red lights. They usually want to ask my what year and then give me a brief synopsis of their days as a bus owner. Bikers and other bus owners either wave or point in my direction at the sky as I pass them on the road.
I had been looking for a good riding hat for the bus and found one last week at Sports Authority while looking for a bike helmet. It was in the golf clothes section and it is perfect. It is similar to a panama. At the Saint Marks river Athena found a white feather so I stuck it in and now I ride around with a feathered cap with maximum style.