MGMT - "Electric Feel" Music Video
I thought this was one of the most creative music videos I have seen in a while. What you think? Leave comments please!
Tuesday, August 18, 2009
Monday, August 3, 2009
Website Development, The Do's and Don'ts.
DO use frameworks.
Frameworks is the serious shit. If you're going into any project, or any task, and you're building it around a framework, you're construction process will always go more smoothly, from CSS frameworks (960.gs, Blueprint), to web application frameworks, such as Django (Python), CakePHP (PHP), Rails (Ruby).
You'll be working with an environment that helps you worry less about structure and standards and focus on the development.
DO write validated HTML.
If you aren't doing this, you'll be shamed upon by anyone looking at your source. You should care because search engines care; however, even writing validated html does not necessarily mean you're writing meaningful html.
DO write meaningful HTML.
Doing <h1><img src="/imgs/mylogo.jpg" /></h1>. Well knock it off. To a computer it means the image link is the header of your page. Instead, do <h1>My Site Name</h1> then add some css style to the h1 tag like, say you wanted to replace that with an image, you could add the attributes: display:none; background:url('/imgs/mylogo.jpg'). If you're writing your html and thinking about the design at the same time, you're simply not doing it right. Instead, just write the content out. Use <h1>s, you may use <div> to split up some sections, or you should use <p> to wrap around your text. Then when you're done with that you add the css attributes to the base tags first. Then follow on with classes for specific detailing. A good example of displaying your page header would be scribefire's homepage.
DO be enthusiastic.
If you want your site to do well, then think about it, be creative, think of different ways to improve your site. Think about your readers wants. Try to get feed back and always challenge yourself. If you wanted to do something in ajax, write down specifically what you're trying to achieve and learn it (Trust me it's not that hard). You're living in an information age, if you feel held back by knowledge, you're not taking advantage.
DON'T get too enthusiastic.
But you can't be too enthusiastic, see when you're researching, you're not really constructing anything. The only way for your site to change, or at least it's core attributes can be your doing. And if you spend too much time researching but not enough time developing, you start to wander too far off and never get anything done due to losing a sense of direction. Try to be focused. Write down a requirement and fulfill it. One requirement at a time. So if you need a comment system but want your users to sign in first. Make the user controller first, and create the log in page. By breaking down the things that you can become more focused on a specific task.
DON'T design sites with flash.
Don't use flash for design purposes. You might think it's cute, or cool. It's not, in fact it's fucking annoying, it wastes my memory, Flash is proprietary and it's support outside of Mac and Windows is pretty shit. You make your site unvisitable from my PHONE. If you have a application that you developed, like a video player or something, then by all means, use it. But if putting flash on your site to represent a menu which doesn't do anything but have a couple hyper links then why are you using flash? Flash is not fast, flash isn't truly re sizable. Flash is one of those things that you should really want to use less of. Adobe hyped you kids up to make you think it was the shit, it really isn't.
Frameworks is the serious shit. If you're going into any project, or any task, and you're building it around a framework, you're construction process will always go more smoothly, from CSS frameworks (960.gs, Blueprint), to web application frameworks, such as Django (Python), CakePHP (PHP), Rails (Ruby).
You'll be working with an environment that helps you worry less about structure and standards and focus on the development.
DO write validated HTML.
If you aren't doing this, you'll be shamed upon by anyone looking at your source. You should care because search engines care; however, even writing validated html does not necessarily mean you're writing meaningful html.
DO write meaningful HTML.
Doing <h1><img src="/imgs/mylogo.jpg" /></h1>. Well knock it off. To a computer it means the image link is the header of your page. Instead, do <h1>My Site Name</h1> then add some css style to the h1 tag like, say you wanted to replace that with an image, you could add the attributes: display:none; background:url('/imgs/mylogo.jpg'). If you're writing your html and thinking about the design at the same time, you're simply not doing it right. Instead, just write the content out. Use <h1>s, you may use <div> to split up some sections, or you should use <p> to wrap around your text. Then when you're done with that you add the css attributes to the base tags first. Then follow on with classes for specific detailing. A good example of displaying your page header would be scribefire's homepage.
DO be enthusiastic.
If you want your site to do well, then think about it, be creative, think of different ways to improve your site. Think about your readers wants. Try to get feed back and always challenge yourself. If you wanted to do something in ajax, write down specifically what you're trying to achieve and learn it (Trust me it's not that hard). You're living in an information age, if you feel held back by knowledge, you're not taking advantage.
DON'T get too enthusiastic.
But you can't be too enthusiastic, see when you're researching, you're not really constructing anything. The only way for your site to change, or at least it's core attributes can be your doing. And if you spend too much time researching but not enough time developing, you start to wander too far off and never get anything done due to losing a sense of direction. Try to be focused. Write down a requirement and fulfill it. One requirement at a time. So if you need a comment system but want your users to sign in first. Make the user controller first, and create the log in page. By breaking down the things that you can become more focused on a specific task.
DON'T design sites with flash.
Don't use flash for design purposes. You might think it's cute, or cool. It's not, in fact it's fucking annoying, it wastes my memory, Flash is proprietary and it's support outside of Mac and Windows is pretty shit. You make your site unvisitable from my PHONE. If you have a application that you developed, like a video player or something, then by all means, use it. But if putting flash on your site to represent a menu which doesn't do anything but have a couple hyper links then why are you using flash? Flash is not fast, flash isn't truly re sizable. Flash is one of those things that you should really want to use less of. Adobe hyped you kids up to make you think it was the shit, it really isn't.
Friday, June 12, 2009
Firefox 3.5 is amazing.
I know my title seems like it's lacking creativity, but I was just being honest. I had to go through 5 minutes of instructions on how to install the firefox 3.5 package for Ubuntu but it wasn't too troublesome. You just needed to goto Mozilla's personal repository and then follow the instructions from there on.
After installing you'll notice not much has change. In fact, everything but the new tab button seems pretty much the same.
But under the hood is where the new firefox 3.5 just shines.
No more render hick-ups, no more rough scrolling, everything is just fast.
Web pages render incredibly fast. And from what I've heard, Firefox implements state of the art standards from the world wide web and is really just a showcase of how powerful the open-source community is.
After installing you'll notice not much has change. In fact, everything but the new tab button seems pretty much the same.
But under the hood is where the new firefox 3.5 just shines.
No more render hick-ups, no more rough scrolling, everything is just fast.
Web pages render incredibly fast. And from what I've heard, Firefox implements state of the art standards from the world wide web and is really just a showcase of how powerful the open-source community is.
Monday, May 18, 2009
DotA HK logo
So I woke up this morning, and I normally sleep with either my girlfriend in my bed or my laptop in my bed. I can't really say which I prefer sleeping with more, but today I woke up thinking about DotA.HK, now I'm a guy who always has a idea but sometimes never takes the time to fully implement them so I decided to take a look at my past work on various web projects, turns out I had some pretty cool stuff ready made for the DotA.HK logo. Take a look!
Monday, May 4, 2009
Personas for Firefox
Don't really like using themes to change the look of your Firefox?
Want something refreshing? Try Mozilla's new personas!
It's a very cool Firefox add on which allows you to easily change your browser's "clothes".
Want something refreshing? Try Mozilla's new personas!
It's a very cool Firefox add on which allows you to easily change your browser's "clothes".
Monday, April 27, 2009
JGhostInfoceptor project started, and open source.
Yes, I've decided to open-source my GhostInfoceptor and have disbanded the original GhostInfoceptor which was written in Python and renamed it as JGhostInfoceptor written in (yep you guessed it) Java.
So far I'm very pleased with how Java deals with buffers and it's garbage collection. It parses the log file very quickly and I believe it could be quite easy to make this application multi-threaded if I needed to.
Anyways I am using GIT for the distributed version control system as I really like it's philosophy and helps me keep track of everything quite easily.
You're all welcome to fork my repository and start making your improvements to propose and I will gladly take the time to consider third-party improvements.
JGhostInfoceptor Repository link
So far I'm very pleased with how Java deals with buffers and it's garbage collection. It parses the log file very quickly and I believe it could be quite easy to make this application multi-threaded if I needed to.
Anyways I am using GIT for the distributed version control system as I really like it's philosophy and helps me keep track of everything quite easily.
You're all welcome to fork my repository and start making your improvements to propose and I will gladly take the time to consider third-party improvements.
JGhostInfoceptor Repository link
Sunday, April 19, 2009
Quake is back, play for free!
I just recently discovered Quake Live where you can play Quake straight from your web browser, not only is it a very solid implementation of the game, it's extremely fast and I absolutely have zero problems getting into the game and playing.
I suggest you all give it a try if you were ever into Quake!
I suggest you all give it a try if you were ever into Quake!
Friday, April 17, 2009
DOTA.HK in trouble with Warden, exit Python, enter Java.
Well it's not secret that we've run into problems at dota.hk getting out hosting platform compliant with Warden. In fact none of us at dota.hk are developers for Ghost++, which is okay, Ghost++ itself has a pretty big pool of developers willing to help out. So support for Warden should be in by the end of the month. I was never scared that we would not find a solution, in fact I'm very happy that Blizzard has done something to fight against Maphack once and for all.
I know for fact Blizzard has nothing against bots that only enhance the experience for users on Battle.net, in fact I remember during the Diablo and Starcraft days that Blizzard actually came out with a source code for developers to develop their own chat bots (this was before we even needed cd-keys to have a chat bot on battle.net).
Right before this happened though I was working on an application I wrote in Python that was finished but was very hard to deploy on multiple servers because of my poor application design. Basically I had to edit a lot of variables within the script for each server.
Example:
NTTHK: config_name = 1Last_update
HKNET: config_name = 2Last_update.
Gosh was that annoying. I wasn't really concerned about the application design, just so long as it worked. Well I'm through with that and have come to a conclusion that using Python to write server-side applications is not the greatest idea, well maybe if it was one server, but not if I was writing for 6 servers that also probably needed to be updated from time to time.
Another problem with Python is that there is no decent IDE's out for it, you have no idea how annoying it is keeping track of indentions. This is how I immediately fell in love with Java+Eclipse. Java and Python are both cross platform which is very important for me as a Linux user in a Windows world. However I think the Java virtual machine is much better than Python's interpreter, because it's very hard to write a application which has problems because of how well the debug system in Java is. I don't have to worry about installing the MySQLdb module, SQLite module on every damn server and Java's C++ (OOP) style syntax makes writing huge applications much more readible and forces good programming practice.
Modular programming in Python is also a huge pain in the ass because you need to define the document folder for where you want to import your custom modules from, this varies from server to server and as far as I know there's no way to automatically determine a path. Don't get my wrong, I love python; but I guess all languages have their ups and downs.
So in a way, I'm relieved that our bots aren't functioning correctly at the moment. It gave me some time to look into using another language to write the application I needed to write. It gave me time to just take a break from all this development and just do some research which was a relief. I guess the biggest task that developers must face is staying on top of their game by constantly forcing themselves to learn new things.
So for the time being, I did some further database optimizations to DOTA.HK V2 to extend it's lifespan, because it looks like we're going to have to delay the release of DOTA.HK v3.
I know for fact Blizzard has nothing against bots that only enhance the experience for users on Battle.net, in fact I remember during the Diablo and Starcraft days that Blizzard actually came out with a source code for developers to develop their own chat bots (this was before we even needed cd-keys to have a chat bot on battle.net).
Right before this happened though I was working on an application I wrote in Python that was finished but was very hard to deploy on multiple servers because of my poor application design. Basically I had to edit a lot of variables within the script for each server.
Example:
NTTHK: config_name = 1Last_update
HKNET: config_name = 2Last_update.
Gosh was that annoying. I wasn't really concerned about the application design, just so long as it worked. Well I'm through with that and have come to a conclusion that using Python to write server-side applications is not the greatest idea, well maybe if it was one server, but not if I was writing for 6 servers that also probably needed to be updated from time to time.
Another problem with Python is that there is no decent IDE's out for it, you have no idea how annoying it is keeping track of indentions. This is how I immediately fell in love with Java+Eclipse. Java and Python are both cross platform which is very important for me as a Linux user in a Windows world. However I think the Java virtual machine is much better than Python's interpreter, because it's very hard to write a application which has problems because of how well the debug system in Java is. I don't have to worry about installing the MySQLdb module, SQLite module on every damn server and Java's C++ (OOP) style syntax makes writing huge applications much more readible and forces good programming practice.
Modular programming in Python is also a huge pain in the ass because you need to define the document folder for where you want to import your custom modules from, this varies from server to server and as far as I know there's no way to automatically determine a path. Don't get my wrong, I love python; but I guess all languages have their ups and downs.
So in a way, I'm relieved that our bots aren't functioning correctly at the moment. It gave me some time to look into using another language to write the application I needed to write. It gave me time to just take a break from all this development and just do some research which was a relief. I guess the biggest task that developers must face is staying on top of their game by constantly forcing themselves to learn new things.
So for the time being, I did some further database optimizations to DOTA.HK V2 to extend it's lifespan, because it looks like we're going to have to delay the release of DOTA.HK v3.
Subscribe to:
Posts (Atom)
