that being said, i just wanted to detail some of the SEO stuff i already know, so that i can have a record of whether or not i’m really learning anything at the SES conference i’m going to next friday. most of this stuff i’ve learned from my buddy arin, who is the lead engineer at a company called fuzz.com, founder at blip.fm, guitarist with suspiciously faustian strumming skills, and honorary ‘king of kong’, among other things… he taught me a bunch of SEO stuff a while ago.

submit your sitemap to yahoo, google, and d-moz

create a sitemap, or an organized list of all the pages on your site. webmaster tools from google and yahoo to submit this sitemap. you can also enter your site into the DMOZ open directory project. doing these things will get your site on the radar, and is one of the first things to do if you aren’t getting any search related traffic.

semantic markup

basically what this means is that when you write your content, you should encode it in a way that makes sense both to human users and web crawlers. when a web spider look at page content, they rank the relevance of your page for search terms based on the context in which those terms are used on your page. having semantic hierarchy between page elements makes your page easier to index, and makes keywords more visible to spiders. If you want to optimize your page for say… “SEO Tips” here are some key elements you should include:

1. title tags: e.g. <title>SEO Tips an Tricks</title>
   
   1. probably the #1 factor in relevance of your page to the keywords you are trying to optimize for.
2. h1 tags: e.g. <h1>10 Best SEO Tips</h1>
   
   1. another very useful persuasion tool in convincing web spiders of the relevance of your page to the search results.
   2. don’t believe me? do a search for a chosen word, then view the source of the first website in your search results. in the window that pops up, do a “find” for “h1″, and you’ll notice 9 times out of 10, the word you searched for appears in h1 tags.
3. URL’s
   
   1. the URL is another place where you can sneak in keywords. instead of http://www.gregdoolittle.com/?xjfga=153425 replace those get variables with something that is readable by a human, which in turn will make the pages seen by crawlers who are trying. a good alternative for this example would be: http://www.gregdoolittle.com/?topic=seo
   2. even better than replacing the get variables with relevant terms, would be using mod_rewrite or isapi_rewrite to configure your site to allow the same URL to be written as: http://www.gregdoolittle.com/topic/seo/. search engines prefer sites that are not dynamically-generated, and having URL’s that appear to be serving actual files instead of database entries will give you a few bumps up in rankings.
4. valid code
   
   1. there are several free tools out there to validate your code. my favorite, is from the world wide web consortium. you can paste in a url, and it will give you a list of all the errors on your page. fixing these errors will make your page content more easily indexed and increase the likelihood of your pages being ranked for specific keywords that may have previously been un-indexable because of the errors on your page.
5. use keywords in your anchor text
   
   1. when you provide links on your site, choose the anchor text carefully, and use keywords if possible instead of meaningless text like the words: “click here”.
   2. Compare the two sentences:
      [Visit our website by clicking here] vs. [Visit our website]
      The latter gives meaning to the anchor, acting as a keyword reference for the target of the link. Search engines know this is more meaningful than the former (”by clicking here”). Your users will appreciate this as well.

optimize the site for download efficiency

not only will your users be happy with cleaner, snappier code, but search engines also favor sites with shorter load-times. having a quick loadtime tells the search engine that it can count on you, increasing the ranking value of your page – longer load-times devalue your page rank.

avoid iframes

in the eyes of the search engine, iframes make your content appear disjointed. spiders look at the relation of one keyword to another within your pages. since web crawlers don’t look at the contents of the iframe while looking at the containing page, the page content will not be indexed. only use iframes if it is required because ajax is giving XSS warnings.

robots.txt

if you haven’t already, do some reading about web crawler standards. robotstxt.org will show you how to create your robots.txt file, add robots meta tags to your site template, and also give you more pointers about how to optimize your site.

seo-friendly background images

this is not a recommendation as a general practice, but only for those rare occasions that call for special fonts. there is some debate whether or not this practice is ethical or alllowable, but david shea (of the css zen garden fame…) supports it, and he has always been a good source of information. basically you use the background image of text, and then pull the inline text out of view, so that users only see a background image. this is something i will try to find a solid answer to at the conference.

inbound links are still hot and sexy, according to me.

setting up accounts for your company on social networking sites (twitter, flickr, linkedin, facebook, myspace), or other sites that are regularly indexed by search engines is a good way to build up an array of inbound links. as your site grows, hopefully, visitors will be providing you with inbound links because they’re bragging about how awesome you are in forums or other peoples blog. providing users with a code-based widget (a picture they make on your site, with a link embedded back to your domain) which they can post wherever they please is a good way to bring in traffic, and increases the number of inbound links you have. while this doesn’t directly optimize for specific keywords, it increases your reputation in the eyes of the search engine, which means it is expecting content from you, and will look to you more frequently when it does its web indices… (i will also attempt to verify this at the SES conference)

there’s a lot of other stuff you can do, like adding images with ‘keyworded’ alt text, installing a blog on your site, etc. but what it comes down to is whether or not you are writing good content. if you write solid content, you only need to make it so some people can find it. once that is accomplished, your rankings will improve on their own. no amount of gaming the system is going to complete your business model — SEO is just a tool for getting your site discovered.

additional resources for SEO best practices

• 10 of the 52 Top SEO Tips
• 10 SEO Tips
• Search Engine Optimization Tips
• SEO - Search Engine Optimization Tricks
• Wordpress SEO Checklist

while i worked at this unnamed company, i was too busy with school to really look for another job, but after about 2.5 months, a former employer called me to ask if i could help them get back on their production schedule. i happily agreed, and worked for my old boss until april - at which point i decided to move to a company in san francisco, Credit.com. there are no investors - we actually make money! (i guess that disqualifies us from calling ourselves “web 2.0″, hehe…). the company focuses on providing information on personal finance, and selling consumer credit services to our users. i work closely with four java engineers. i am the lead front-end developer - which basically means i am responsible for UI/UE, and all issues around the user-facing code.

there’s a lot of awesome people here. good compensation and fringe benefits, including working from home once per week, and free massages on fridays. it’s like a 1990’s start-up in here… except that we’re actually making money. they’re sending me to the SES conference in san jose, august 22nd. i resisted at first because i feel like i have a pretty good grasp of SEO already, but hey - it’ll be a good field trip, and i admit, there’s still plenty of SEO stuff to learn…

hopefully i can detail my SEO knowledge here on my blog before i go to this conference, so that i can easily see whether or not i end up learning enough to make the conference worth its entry fee…. (so check back next week for that, as well as my notes from the conference a fortnight from now (yes bitches, ‘a fortnight’)).

starting at cafe boronne in menlo park -> santa cruz ave. -> sand hill road -> the eastern face of the mountain on old la honda -> north on skyline to 84 -> down the west side of 84 towards the coast -> up the western side of old la honda back to skyline -> same stretch of skyline back to 84 again -> down the east side of the mountain down the hill back towards home -> portola road -> alpine rd. -> santa cruz ave. -> back to boronne’s.

34.7 miles, over 3300 vertical gain – my longest ride to date.

Over the weekend, I started and finished a freelance design project for a friend over at carbonharmony.com. This week, I start my new position with Secorix Inc. as a Software Engineer / QA Support Engineer / Designer / Programmer.

Which brings me to the point of this post: I’m going to be working for Secorix as an independent contractor. I’ve been reading various info about working as a contractor, and I’ve talked to everyone I know who already is one… and it looks like the simplest and cheapest way to address my needs would be to form a sole proprietorship. At some point during my first day, I’ll ask other employees what they think is the best route to take. Then hopefully i’ll be meeting with a CPA, registering my business with the county, and setting up a business account at Wells Fargo. Even if it turns out that I don’t need to set up a business for this new job, I may do it anyway so that I can work as a full-fledged freelancer.

I first heard about OCW on NPR. The story reported on how MIT wanted to provide its course offerings to the rest of the world for free, as part of a community outreach program. Several students were upset because they didn’t like the idea of sharing their expensive education with the rest of the world. Other students put forth the argument that by giving the courses out to the rest of the world, the worldwide level of flourishing would increase, thereby increasing the individual’s ability to flourish. On the OCW site, you can read their guiding principal:

MIT is committed to advancing education and discovery through knowledge open to everyone.

It is a generous thing to do, and so far I am amazed by the course offerings. There is so much information available – it’s overwhelming. I’ve looked at several different Foreign Language, Pre-Law, and Computer Science classes. If you’re a do-it-yourself learner, I highly recommend looking into OCW. While sites like OED and Wikipedia offer tons of information, OCW actually helps you learn the subject matter in an efficient manner. If you really want to become an expert on a subject area, OCW is a great tool to get you there.

i figured out a cool way to use spirals to create these great kaleidoscope images. enjoy!

see the flash movies

see the flash movie

my boss is now asking me to work with him on building social network software for a conference focused on internet security. i told him about some of the open source social network aps that you can download/customize/install on your own server, but it looks like he wants to do something a bit more home-cooked. if this works out it could be a really amazing project, but due to the sheer scale, i’m a little nervous about it. i’ll let you know how it works out.

http://developer.yahoo.com/yui/theater/

first of all, it’s free (at least right now it is). i don’t know how long it’s been since at&t stopped providing the service, it seems like almost 10 years. on a cellphone, the cost for a phonebook service had been in the 1$ range.

second of all, the voice recognition feature is really, really good. i’ve compared to both bmw’s idrive recognition and apple’s built-in finder voice recognition on OSX.4. it’s hands down the best voice recognition technology i’ve ever experienced. if it’s not sure what you said, i.e. doesn’t have an exact match, it gives you a list of things you may have said, and you can confirm by number.

the computerized voice is better than anything i’ve ever seen. very natural sounding. the intonation makes it very easy to understand. there is some room for improvement here, but it is by far the best ever computer voice at this point in time.

the interface: you say the city and state; you confirm the location; you say which listing you need; you then get the address and phone number; you’re prompted as to whether or not you’d like the info in a text message; it then repeats the info; it then connects you automatically to the phone number. there’s nothing more i could want from a 411 service. it is brilliant.

really, you ought to try it out for yourself. go on, pick up your phone… (1.800.GOOG.411)

what’s new at TechWeb, you ask?

Our MediaLive group (the group I work with) recently took control of the software 2008 conference, a product of the people at sandhill.com. we’ve also expanded several of our current shows to international venues (Interop São Paulo, Web 2.0 Berlin…). We also recently launched bMighty.com, a journal/blog devoted to SMB’s.

I’ve been working with OpenAds, configuring our own ad server to replace our paid contract with double-click. I’m also trying to get us off of HitBox, which could easily be replaced with Google Analytics. I’ve installed the analytics-javascript file on all of our templates, now I just have to convince the higher-ups that the statistics from on Google Analytics are as good if not better than what we get from HitBox. Ah… bureaucracy…

current projects:

• redo my php-craps game to take advantage of ajax, also integrate a gambling/wager feature, and beef up the graphics a bit
• create simplified ecommerce wedding registry
• negotiate a redesign

you can see what i did here, and hopefully soon, you will see it on www.carbonharmony.com.

The interface is 100% more lick-able, the colors are brighter and bolder, and the code is much more clean. As soon as I get an opinion on it, I’ll fix it up and put the files at the root of the site, followed by carrying the re-design through to other pages of the site.

before | after

When we started using a flash slideshow on one of our homepages, we were in a scramble to get all the copy-edits and registration systems up and running. Now we have some time to breath, so I went back and found an article about out how to get around this problem. (Read it here)

Adding a few lines of javascript takes care of the click-to-activate problem.

objects = document.getElementsByTagName("object");
for (var i = 0; i < objects.length; i++)
{
    objects[i].outerHTML = objects[i].outerHTML;
}

You can see it on one of our sites here: www.interop.com

***update 6/26:

Noticed some conflicts caused by the above version. Mainly happening with our drop down menus, and the javascript we use in our ad network didn’t like having it’s “outerHTML” messed with. The version below works great thus far:
http://activecontent.blogspot.com/

http://www.splintered.co.uk/experiments/49/

basically all you need to do is add this to your css, and you’ll get scrollbars on every page:

html { min-height: 100%; margin-bottom: 1px; }