Chances are, someone in the community has experienced your very problem, and it's likely someone has a solution already. At the very least, you'll get some ideas from people who actually use the product.
I doubt the brain is telling the fingers to spread to an absolute unit of length, but rather relative to some object. In both cases, just large enough to grasp the circular object, however large it might be.
I know the point he's making; I read the blog post. I don't know why you assume I didn't. I'm saying that I believe it is unrealistic to expect that a user will not attempt to interact with a notification on the screen if it's there. More than likely they will be entirely confused when the mouse goes "under" the notification, which is unlike anything else on the system.
His point is that the ability to interact with notifications means that you have to decide, in the two or three seconds that the notification is on-screen, whether or not you want to interact with it. That's distracting, which makes notifications annoying.
You don't have to decide anything, actually. You can read it or not. Having the ability to interact with the notification doesn't change anything about the interaction with the notifications up until the point the user decides to interact with it, which I believe the user will do initially anyway regardless of whether he can actually interact with it or not.
Personally, I'm with him. I hate it when I try to switch desktops, close a window, open my IM client, or whatever, and a notification pops up to block my click (or catch it and do something unexpected).
This is a valid concern that can be addressed in other ways, but I don't think the way highlighted in the article is a good one because, again, users will not expect that their mouse will go under the notification. I guarantee a non-insignificant number will sit and wait for the notification to go away before clicking on something they want underneath. This is indicative of an interface that is not as intuitive as it could be.
Having the notification bubbles disappear when you mouse over (well, under) them doesn't seem usable. The user will see the bubble and want to interact with it in some way. Mousing over should decrease opacity and allow the user to interact with the dialog, such as immediately remove it or click on it to bring up the application that spawned the notification. I'm very familiar with computers, and it still seems very strange to "mouse under" something.
There are already a few solutions for bug tracking in a git repository (here is one example that also includes a minimalist web interface), though from what I've seen nothing really compares to hosted solutions as of yet. It's great for simple projects, though, especially when you're just starting out and want to keep track of todo items and such.
Confirmed. It's played everything I've thrown at it, and it is also considerably faster. With NSPluginWrapper, when I loaded a page full of flash graphs, the browser became sluggish for some time. With the alpha, the graphs load up instantly. So far I'm very impressed.
I read Wordcraft recently, and was highly entertained. It talks about the process of naming and follows a few names in detail, including BlackBerry and a few others.
In the current environment of massive liquidation, sure, we're in a deflationary period if anything. But what happens once the market is sufficiently deleveraged given that all of the world's banks have been printing money non-stop? Historically speaking, the answer is inflation.
Have you tried to sit down and create a font before? Making a font that is suitable for a certain purpose, be it attractive headings and titles or body copy or whatever, takes a very long time, a lot of hard work, a lot of know how, and yes, artistic talent. DRM on a font is no less absurd than DRM on software, music, movies, photos, or the like.
Now, that's not to say that DRM has a place in web fonts, and that's certainly not to say that EOT is the way to implement it. Comparing a font to a color, though, is the utmost of absurdity.
In addition, even though we HAVE gigabit ethernet today, all but the most expensive networking equipment (read: most consumer equipment) can't push that much data through. And even then, it's sufficient for most purposes. Most home built/lower end NAS devices don't even need gigabit as, even with gigabit equipment, they'll push 30MB/s or so on a good day before the CPU, NIC, or router/switch chokes.
It causes consistent behavior once you train it;) I've got Quicksilver on my mac, and it learned soon after I installed it that "f" was Frozen Throne, "ff" was firefox, "t" was textmate, "te" was text edit and so on. It's incredibly valuable. I can see, though, how it would cause a novice user some strife as the behavior would be (at least perceivably) erratic early on.
I worked on Chipmark, which is both an open source project and a school project. It was also one of the most valuable classes I took in college. The grading was lots less concrete than any other class, which I'm sure made it more difficult on the instructor. It also made it hard on students as you don't have a set of concrete requirements to gauge yourself with. Chipmark shows that it is not impossible to do and is in fact extremely valuable.
Plus TSA is incompetent anyway. I accidentally left a 4 inch hunting knife in my carryon from a recent camping trip. Realized it was in my bag only upon reaching my destination.
Solaris 10's Service Management Facility (SMF) currently loads services in parallel. It keeps track of a service's dependencies so it can boot them in the proper order.
It seems to be pretty vast, you're right! I was able to find a Hindemith clarinet sonata I've been looking for for a while... I bought those 4 movements for a grand total of $3.96! What a steal!
If you wanted all that stuff, minus rails, plus PHP it'd be the same difficulty to get set up, though, so I fail to see how this is relevant to the topic at hand. That's my only confusion with this, I guess.
Anyway, as for a setup script, I don't have one. I just compile and configure each individual piece of software by itself. It only needs to be done once, so a script doesn't seem necessary. Maybe I'm misunderstanding what you mean.
I am using WebDAV and SSL with apache. To compile those two in you need enable-dav and enable-ssl as configure flags. This is all in the Apache configure docs. Those plus the ones I mentioned previously for proxy balancer and you should be good to go. SSH is easily set up on debian-like machines with apt-get (apt-get install ssh). Mongrel is installed as a rubygem, and is actually far better than webrick as a testing server.
I wouldn't exactly say the backers aren't referring to it as an OS. Right on the front page it reads, "ajaxWindows is a virtual operating system". They also refer to it as an OS in the demo video.
At least to me, it doesn't sound like you are having problems with Rails but instead with compiling Apache to work with mod_proxy_balancer and getting subversion set up on Ubuntu? Apache can be a pain to configure and compile correctly, but I hardly see how you can fault Rails for that.
Incidentally, your requirements are very similar to mine, and I just today set up almost that exact thing on a dedicated Solaris box. Compiling apache was the most "difficult" part, but it's just a matter of figuring out what you want in advance and configuring that stuff in. Specifically for proxy balancer you want enable-proxy, enable-proxy-html, and enable-proxy-balancer. This is all in the apache docs. You could also try Nginx, I hear a lot of people have a better experience there when your requirements don't include php and the like and instead you are just proxying to mongrels or something.
The only configs relevant to a rails application should be those in your config directory. Everything else is in the realm of system administration, in my opinion. Your qualms seem mostly with compiling specific software on a linux platform.
It is completely fucking ridiculous to get a server set up that you could seriously test with or, God forbid, move into production.
That's strange... one thing I really like about rails is how it's extremely easy to load up a test server with either Mongrel, lighttpd, or *shudder* if you must, Webrick. The benefit of using mongrel is that it's production ready with the "pack-o-mongrels" approach.
As someone who has deployed a few Rails-based web applications in production, I simply can't fathom what your problems are a result of. I've never had to edit obscure configuration parameters - I don't even know where you'd go to find them. Perhaps you could share with us some of your specific problems? What did you have to configure, etc? Why doesn't "script/server" as a testing server (using Mongrel, Lighttpd, or Webrick) work for you?
Slashdot Mirror
Chances are, someone in the community has experienced your very problem, and it's likely someone has a solution already. At the very least, you'll get some ideas from people who actually use the product.
I doubt the brain is telling the fingers to spread to an absolute unit of length, but rather relative to some object. In both cases, just large enough to grasp the circular object, however large it might be.
I know the point he's making; I read the blog post. I don't know why you assume I didn't. I'm saying that I believe it is unrealistic to expect that a user will not attempt to interact with a notification on the screen if it's there. More than likely they will be entirely confused when the mouse goes "under" the notification, which is unlike anything else on the system.
You don't have to decide anything, actually. You can read it or not. Having the ability to interact with the notification doesn't change anything about the interaction with the notifications up until the point the user decides to interact with it, which I believe the user will do initially anyway regardless of whether he can actually interact with it or not.
This is a valid concern that can be addressed in other ways, but I don't think the way highlighted in the article is a good one because, again, users will not expect that their mouse will go under the notification. I guarantee a non-insignificant number will sit and wait for the notification to go away before clicking on something they want underneath. This is indicative of an interface that is not as intuitive as it could be.
Huh? It sounds nothing like that. If you want to draw a comparison, it looks and sounds a lot like Growl notifications on OS X.
What I meant to say was, "increase opacity."
Having the notification bubbles disappear when you mouse over (well, under) them doesn't seem usable. The user will see the bubble and want to interact with it in some way. Mousing over should decrease opacity and allow the user to interact with the dialog, such as immediately remove it or click on it to bring up the application that spawned the notification. I'm very familiar with computers, and it still seems very strange to "mouse under" something.
Here's the exploit code referenced in the article update... The second one apparently works on Vista, too. http://www.milw0rm.com/exploits/7403 http://www.milw0rm.com/exploits/7410
There are already a few solutions for bug tracking in a git repository (here is one example that also includes a minimalist web interface), though from what I've seen nothing really compares to hosted solutions as of yet. It's great for simple projects, though, especially when you're just starting out and want to keep track of todo items and such.
Pretty sure I had to edit the mysql conf to allow external connections as well, but I'm sure it depends on the distribution (in my case, Debian).
Confirmed. It's played everything I've thrown at it, and it is also considerably faster. With NSPluginWrapper, when I loaded a page full of flash graphs, the browser became sluggish for some time. With the alpha, the graphs load up instantly. So far I'm very impressed.
I read Wordcraft recently, and was highly entertained. It talks about the process of naming and follows a few names in detail, including BlackBerry and a few others.
In the current environment of massive liquidation, sure, we're in a deflationary period if anything. But what happens once the market is sufficiently deleveraged given that all of the world's banks have been printing money non-stop? Historically speaking, the answer is inflation.
Have you tried to sit down and create a font before? Making a font that is suitable for a certain purpose, be it attractive headings and titles or body copy or whatever, takes a very long time, a lot of hard work, a lot of know how, and yes, artistic talent. DRM on a font is no less absurd than DRM on software, music, movies, photos, or the like.
Now, that's not to say that DRM has a place in web fonts, and that's certainly not to say that EOT is the way to implement it. Comparing a font to a color, though, is the utmost of absurdity.
In addition, even though we HAVE gigabit ethernet today, all but the most expensive networking equipment (read: most consumer equipment) can't push that much data through. And even then, it's sufficient for most purposes. Most home built/lower end NAS devices don't even need gigabit as, even with gigabit equipment, they'll push 30MB/s or so on a good day before the CPU, NIC, or router/switch chokes.
It causes consistent behavior once you train it ;) I've got Quicksilver on my mac, and it learned soon after I installed it that "f" was Frozen Throne, "ff" was firefox, "t" was textmate, "te" was text edit and so on. It's incredibly valuable. I can see, though, how it would cause a novice user some strife as the behavior would be (at least perceivably) erratic early on.
Something similar happened with the 10.4.11 update as well.
I worked on Chipmark, which is both an open source project and a school project. It was also one of the most valuable classes I took in college. The grading was lots less concrete than any other class, which I'm sure made it more difficult on the instructor. It also made it hard on students as you don't have a set of concrete requirements to gauge yourself with. Chipmark shows that it is not impossible to do and is in fact extremely valuable.
Plus TSA is incompetent anyway. I accidentally left a 4 inch hunting knife in my carryon from a recent camping trip. Realized it was in my bag only upon reaching my destination.
Solaris 10's Service Management Facility (SMF) currently loads services in parallel. It keeps track of a service's dependencies so it can boot them in the proper order.
It seems to be pretty vast, you're right! I was able to find a Hindemith clarinet sonata I've been looking for for a while... I bought those 4 movements for a grand total of $3.96! What a steal!
So install mongrel then. If you have mongrel installed, script/server runs mongrel by default :)
If you wanted all that stuff, minus rails, plus PHP it'd be the same difficulty to get set up, though, so I fail to see how this is relevant to the topic at hand. That's my only confusion with this, I guess.
Anyway, as for a setup script, I don't have one. I just compile and configure each individual piece of software by itself. It only needs to be done once, so a script doesn't seem necessary. Maybe I'm misunderstanding what you mean.
I am using WebDAV and SSL with apache. To compile those two in you need enable-dav and enable-ssl as configure flags. This is all in the Apache configure docs. Those plus the ones I mentioned previously for proxy balancer and you should be good to go. SSH is easily set up on debian-like machines with apt-get (apt-get install ssh). Mongrel is installed as a rubygem, and is actually far better than webrick as a testing server.
I wouldn't exactly say the backers aren't referring to it as an OS. Right on the front page it reads, "ajaxWindows is a virtual operating system". They also refer to it as an OS in the demo video.
At least to me, it doesn't sound like you are having problems with Rails but instead with compiling Apache to work with mod_proxy_balancer and getting subversion set up on Ubuntu? Apache can be a pain to configure and compile correctly, but I hardly see how you can fault Rails for that.
Incidentally, your requirements are very similar to mine, and I just today set up almost that exact thing on a dedicated Solaris box. Compiling apache was the most "difficult" part, but it's just a matter of figuring out what you want in advance and configuring that stuff in. Specifically for proxy balancer you want enable-proxy, enable-proxy-html, and enable-proxy-balancer. This is all in the apache docs. You could also try Nginx, I hear a lot of people have a better experience there when your requirements don't include php and the like and instead you are just proxying to mongrels or something.
The only configs relevant to a rails application should be those in your config directory. Everything else is in the realm of system administration, in my opinion. Your qualms seem mostly with compiling specific software on a linux platform.
That's strange... one thing I really like about rails is how it's extremely easy to load up a test server with either Mongrel, lighttpd, or *shudder* if you must, Webrick. The benefit of using mongrel is that it's production ready with the "pack-o-mongrels" approach.
As someone who has deployed a few Rails-based web applications in production, I simply can't fathom what your problems are a result of. I've never had to edit obscure configuration parameters - I don't even know where you'd go to find them. Perhaps you could share with us some of your specific problems? What did you have to configure, etc? Why doesn't "script/server" as a testing server (using Mongrel, Lighttpd, or Webrick) work for you?