I am using the new website to get to grips with CSS layouts. I know a fair bit about CSS although I've never really made a full layout for a site, and it's proving harder than I thought.
I used a layout provided by 'dynamic drive, thinking that it would work in all browsers no matter what. I was wrong. So I figure I'm going to do the layout myself, which may take longer than I thought.
New ideas are coming to me all the time but I'm trying to focus on Keeping it simple. The logo has been designed although I still can't publish it as I only want people to see the site when it is finished, it is too unique for people to see so soon.
I'm essentially going to use the site to fund University as well as xoog, I don't expect too much from it but I do see the potential.
Friday, 29 December 2006
Tuesday, 26 December 2006
new website idea
i have thought of a new website idea that could potentially be as big as milliondollarhomepage.
its a unique idea that i havent seen before online and after a few searches, i found nothing like it.
im going to be building it as soon as i have finished a few mini-projects on xoog.
if im right, and i usually am, marketing of the site will be completely viral, and it will also be a good way of funding some larger projects for xoog; new cameras, etc.
ill post a link when it is finished, most likely in a few weeks (and hopefully when the seo diary might have a couple of subscribers)
its a unique idea that i havent seen before online and after a few searches, i found nothing like it.
im going to be building it as soon as i have finished a few mini-projects on xoog.
if im right, and i usually am, marketing of the site will be completely viral, and it will also be a good way of funding some larger projects for xoog; new cameras, etc.
ill post a link when it is finished, most likely in a few weeks (and hopefully when the seo diary might have a couple of subscribers)
optimising for google sitemaps
A big topic in the forums is how people can generate a sitemap when they have dynamically created website such as xoog. this is easy, i dont understand why people have had so much trouble.
The route I have gone down is to submit a text file listing each page on a new line.
This isnt the best way of doing things, although google indexes the pages, so as far as im concerned, my way is perfectly fine.
What you do is go into sitemaps and tell google where your text file is located, and it spiders the text file so it knows what pages you have.
googlebot doesnt seem to spider as deep as some other search engines, right? i think it just gets bored. but i still need it to find every unique page of mine.
My Method
what i do is use REQUEST_URI on every page that the user visits and log it into a database table. I make sure that every entry is unique.
I then use a combination of php to output a file, and htaccess to make the file appear as a .txt (or .xml if you swing that way) file.
this way my users (and the deeper-crawling bots) generate my sitemap file for me :) Plus every time google checks my sitemap file it will always be updated.
This method relies on users and bots clicking away through your site, however if they arent doing this... Is there any point in your site ever being indexed in google anyway? No!
The route I have gone down is to submit a text file listing each page on a new line.
This isnt the best way of doing things, although google indexes the pages, so as far as im concerned, my way is perfectly fine.
What you do is go into sitemaps and tell google where your text file is located, and it spiders the text file so it knows what pages you have.
googlebot doesnt seem to spider as deep as some other search engines, right? i think it just gets bored. but i still need it to find every unique page of mine.
My Method
what i do is use REQUEST_URI on every page that the user visits and log it into a database table. I make sure that every entry is unique.
I then use a combination of php to output a file, and htaccess to make the file appear as a .txt (or .xml if you swing that way) file.
this way my users (and the deeper-crawling bots) generate my sitemap file for me :) Plus every time google checks my sitemap file it will always be updated.
This method relies on users and bots clicking away through your site, however if they arent doing this... Is there any point in your site ever being indexed in google anyway? No!
the niche
I have to admit - The keywords I am optimising xoog for are easy. They have quite a large niche, and the competition isnt that big for the keywords. The biggest competition I have for keywords is tilllate and the club's/events website themselves.
neither of these have a lot of decent content for google to lay its teeth in. this is where my optimisation comes in.
tilllate does pictures, and doesnt have people who have much knowledge in SEO and content optimisation. However due to the fact that they are a lot more established than xoog, they rank higher for a lot of words... Although the gap is and HAS been closing down extremely fast.
based on some well known keyword tools I have been optimising my content.
At first I believed that my visitors only need to come from the business cards I hand out to everybody, however more recently I realised that a website, whathappenedlastnight.net gets a LOT of visitors just because of its content - they have a lot of written content and no longer hand out business cards. I am going to take WHLN's visitors in time. Based on Alexa stats I estimate that they get about 400 uniques per day through their content.
I'm going to be introducing an article system optimised to keywords, although I need to write more articles before i can officially introduce it. At the moment I am just placing a link to the articles at the bottom so that google at least indexes them for now.
neither of these have a lot of decent content for google to lay its teeth in. this is where my optimisation comes in.
tilllate does pictures, and doesnt have people who have much knowledge in SEO and content optimisation. However due to the fact that they are a lot more established than xoog, they rank higher for a lot of words... Although the gap is and HAS been closing down extremely fast.
based on some well known keyword tools I have been optimising my content.
At first I believed that my visitors only need to come from the business cards I hand out to everybody, however more recently I realised that a website, whathappenedlastnight.net gets a LOT of visitors just because of its content - they have a lot of written content and no longer hand out business cards. I am going to take WHLN's visitors in time. Based on Alexa stats I estimate that they get about 400 uniques per day through their content.
I'm going to be introducing an article system optimised to keywords, although I need to write more articles before i can officially introduce it. At the moment I am just placing a link to the articles at the bottom so that google at least indexes them for now.
Stop stealing my images!
One of the problems with xoog is that so many people are copying and pasting the images and putting them on their myspace profiles. for a while i have decided that this is good, as they dont get rid of my watermark and it means people come to xoog and the brand gets circulated.
more recently i have decided that i want to create my own myspace codes, which allows me to monitor xoog's influence through extremely clever coding - which is too clever to mention publically :-)
I have done a lot of reading and it is pretty well established that webmasters can only make it difficult for users to steal their images, not impossible. this in mind, I am going to make it as hard as possible.
Lots of suggestions have been made in the forums and articles, and from what I see it is important to use a combination of techniques to make it as hard as possible.
Problem: Users stealing images
Solution:
- Prevent users from right-clicking - No! Dont do this.
DO... Put the image in the background of a cell, and then put a stretched transparent .gif/.png for added effect. The average user will be confused that when they successfully save the image to their computer, they will only have a transparent image. extremely off-putting to them.
- Stop users from printing the screen. This can be done in javascript, I
(think) I have found the code to do this although as this isnt a coding blog im going to make you find it yourself.
- Encrypt the source. This only works with javascript enabled, although it prevents people from finding the hidden filename that is in the background of the cell.
- Convert the image to a swf! This can stop copy and pasting, I dont know why people havent mentioned this in their articles or forum posts... For xoog I cannot do this, hnowever, because I need users to be able to click on the image to record certain information.
- Watermarks. A lot of people believe watermarks are the best way to stop the stealing of images. not me. I want my site to be user friendly - how can it be with huge and complex watermarks?? One on the bottom right is fine for me.
more recently i have decided that i want to create my own myspace codes, which allows me to monitor xoog's influence through extremely clever coding - which is too clever to mention publically :-)
I have done a lot of reading and it is pretty well established that webmasters can only make it difficult for users to steal their images, not impossible. this in mind, I am going to make it as hard as possible.
Lots of suggestions have been made in the forums and articles, and from what I see it is important to use a combination of techniques to make it as hard as possible.
Problem: Users stealing images
Solution:
- Prevent users from right-clicking - No! Dont do this.
DO... Put the image in the background of a cell, and then put a stretched transparent .gif/.png for added effect. The average user will be confused that when they successfully save the image to their computer, they will only have a transparent image. extremely off-putting to them.
- Stop users from printing the screen. This can be done in javascript, I
(think) I have found the code to do this although as this isnt a coding blog im going to make you find it yourself.
- Encrypt the source. This only works with javascript enabled, although it prevents people from finding the hidden filename that is in the background of the cell.
- Convert the image to a swf! This can stop copy and pasting, I dont know why people havent mentioned this in their articles or forum posts... For xoog I cannot do this, hnowever, because I need users to be able to click on the image to record certain information.
- Watermarks. A lot of people believe watermarks are the best way to stop the stealing of images. not me. I want my site to be user friendly - how can it be with huge and complex watermarks?? One on the bottom right is fine for me.
Optimising the automated gallery - filenames, alt attributes and title attributes
The Google algorithm seeks out and destroys automated content in its sleep. Its a fact. This is simply because real content is generally useful content.
However, what about xoog where the pictures ARE real content. Because of the nature that the pictures are put online, to Google it looks as if they have simply been automatically placed on the internet. Google will not like this and hence not index all of my images.
Thousands of images are put on xoog every month. It isn't reasonable to write a description for every single one of these pictures. this description can be placed in the title and attribute tags of an image. This causes a problem.
I have figured a potential workaround to this problem, which could potentially be done on every gallery.
Problem: Automation image creation makes it hard to distinguish every image, which results in Google not being able to tell the difference between each image, and also not assigning any keywords to the image.
Resolution: Create a database with keywords related to a topic. My tables are going to be structured in a way that keywords will be assigned 'groups'. This will allow future versatility and allow me to actually relate my images to the appropriate keywords. the groups will be assigned names such as '42nd Street' or 'Mtwo'... After all, I don't want to deceive Google, just help it along.
When the images are being pulled from the file server, and added to the database, the words will be pulled from an the appropriate table. The image will be optimised based on these keywords. For example if the keywords are "42nd street night club manchester"...
filename:
42street_0123.jpeg -> 42street_0123-42nd-street-night-club-manchester.jpeg
alt/title attributes:
"42nd street night club manchester"
the keywords will be a lot more random then that, as they will be randomly taken from the database, and also a random number of keywords will be taken for more combinations.
why will this help?
If each group has 50 key words, and between 5 and 10 words are generated on every image. there are hundreds of thousands of possible keyword results. Every single image in the gallery will be unique. Google Like.
However, what about xoog where the pictures ARE real content. Because of the nature that the pictures are put online, to Google it looks as if they have simply been automatically placed on the internet. Google will not like this and hence not index all of my images.
Thousands of images are put on xoog every month. It isn't reasonable to write a description for every single one of these pictures. this description can be placed in the title and attribute tags of an image. This causes a problem.
I have figured a potential workaround to this problem, which could potentially be done on every gallery.
Problem: Automation image creation makes it hard to distinguish every image, which results in Google not being able to tell the difference between each image, and also not assigning any keywords to the image.
Resolution: Create a database with keywords related to a topic. My tables are going to be structured in a way that keywords will be assigned 'groups'. This will allow future versatility and allow me to actually relate my images to the appropriate keywords. the groups will be assigned names such as '42nd Street' or 'Mtwo'... After all, I don't want to deceive Google, just help it along.
When the images are being pulled from the file server, and added to the database, the words will be pulled from an the appropriate table. The image will be optimised based on these keywords. For example if the keywords are "42nd street night club manchester"...
filename:
42street_0123.jpeg -> 42street_0123-42nd-street-night-club-manchester.jpeg
alt/title attributes:
"42nd street night club manchester"
the keywords will be a lot more random then that, as they will be randomly taken from the database, and also a random number of keywords will be taken for more combinations.
why will this help?
If each group has 50 key words, and between 5 and 10 words are generated on every image. there are hundreds of thousands of possible keyword results. Every single image in the gallery will be unique. Google Like.
A Mild Introduction
I just thought that I might as well give a little introduction to the blog. First of all, this is my first blog, second of all, I'm not too bothered whether or not people read the blog - I'm building it as a personal reference to my findings in the world of SEO.
I don't class myself as an SEO expert, however from the people I've spoken to who claim to be SEO experts themselves, I know a hell of a lot about search engine optimisation.
I have designed xoog.net from scratch, it was originally a project that was meant to be a clone of competitors such as tilllate.com, textmi.com, dontstayin.com, fatsoma.com amongst other similar sites that specialise in nightlife photography.
xoog was originallly intended as a university project, but I was forced to turn it into a proper website after I was fired as a photographer from tilllate.com, which is coincidentally the largest nightlife photography website in the world.
Because of my background, xoog mimics a lot of features from tilllate.com; as when I was designing xoog I did not intend it to go public, and this was not a concern.
It is not my goal to change xoog so it is as different to tilllate as possible; it is a slow process as it is a powerful script, however luckily i designed the entire site so that it could be easily modified in the future at any point.
as my university course does not focus on thing such as search engine optimisation (although it really should), I also didn't think about optimising xoog for the search engines. This has become an interesting challenge to me and it has gave birth to this very blog.
The blog will be about my findings and optimisations of xoog, as well as new features. it is a diary for ME, if you read it, I hope you find my discoveries interesting, there will certainly be a few new ideas brought forth to the SEO world, which will especially be useful for webmasters building web galleries and communities that are similar to xoog.net
I don't class myself as an SEO expert, however from the people I've spoken to who claim to be SEO experts themselves, I know a hell of a lot about search engine optimisation.
I have designed xoog.net from scratch, it was originally a project that was meant to be a clone of competitors such as tilllate.com, textmi.com, dontstayin.com, fatsoma.com amongst other similar sites that specialise in nightlife photography.
xoog was originallly intended as a university project, but I was forced to turn it into a proper website after I was fired as a photographer from tilllate.com, which is coincidentally the largest nightlife photography website in the world.
Because of my background, xoog mimics a lot of features from tilllate.com; as when I was designing xoog I did not intend it to go public, and this was not a concern.
It is not my goal to change xoog so it is as different to tilllate as possible; it is a slow process as it is a powerful script, however luckily i designed the entire site so that it could be easily modified in the future at any point.
as my university course does not focus on thing such as search engine optimisation (although it really should), I also didn't think about optimising xoog for the search engines. This has become an interesting challenge to me and it has gave birth to this very blog.
The blog will be about my findings and optimisations of xoog, as well as new features. it is a diary for ME, if you read it, I hope you find my discoveries interesting, there will certainly be a few new ideas brought forth to the SEO world, which will especially be useful for webmasters building web galleries and communities that are similar to xoog.net
Subscribe to:
Posts (Atom)