Thinking about it, the ability to get to my home machine from anywhere could be a lifesaver. It also would make carrying a flash drive useless. Although I've used a jump drive for a number of years, Murphy's Law always takes over. It never fails that the file I need is always on my home machine. Plus, I lost a drive once with a good amount of data on it and it forced me to make major security changes.

The Academy utilizes the fast-and-easy GoToMyPC.com. This tool performs wonderfully and is simple to set up. The only drawback is the price to use it. For personal use, it's $19.95/month. I thought this was pretty steep, so I set out to find a more economical way.

After a short amount of hunting, here's what I came up with.

RealVNC and DynDNS
RealVNC.com is a UK-based company spealizing in remote access software. As a test, I downloaded and installed their free Personal Edition. If things worked out, I planning on purchasing their $50 'enterprise edition'. Their package loads a 'VNC Server' and a 'VNC Viewer' application. The server and viewer is a whopping 800KB.

To set up the server, I following their online setup guide. It was pretty easy. What's nice about the package is the Java viewer. That means there's no download; any Java-enabled browser will work.

Configuring Firewall and Router
After the software install, I had to configure my Firewall and Wireless Router to accept connections from the Internet. This has been the toughest part. As I tinkered, I had the idea of accessing my computer via my company's domain name (Grabers.com) instead of just some IP addess. But because I use a cable mdoem for my connection and my IP address changes every so-often, I needed a solution to map my changing IP to my domain and have things refresh as they changed. This is where dyndns.org comes in: dynamic dns services.

Setting up DynDns.org
This part was tricky. You must purchase 'Custom DNS' services from them at $24.95 a year. If you want to forgo this, they also offer a free version, but the domain names you can pick from a limited. I thought it would be 'more enterprise' if I used my own domain.

After purchasing, you must then setup up host-services. This is where it got tricky. I had to setup (or re-create) all of A, MX, and CNAME entries in my Zone File at my hosting provider. This info as easily obtained.

Adjusting Nameservers
The final step in getting this all to work is adjusting your domain name server pairs are your registrar. You enter the DNS entries given by DynDNS.org. The only tedious part is waiting for everything to propogate.

What appears to happen is that requests are directed at DYNDNS. From their, the Hosts file takes over. In my case, all web servering and email is directed back to my hosting provider. But my remote-control services are directed to my personal machine.

By the time everything settled, it was 6:00ish. I had to get me router back working and continue to make adjustments. These will probably be ongoing over the next few days.

Final tweaks have included making things convenient when logging-in. Accomplished via a simple 301 redirect.

The most ironic part was testing. The first time I succesfully logged-in, I used my Gotomypc connection to access my office computer. Weird.

Cross Platform Access
A very cool thing about this method is the ease in which you have a 'virtual cross-platform' machine. When I'm traveling, I bring my Powerbook. But since a majority of my day-to-day stuff is PC related, I now have a way to work on that platform without relying on VirtualPC. I simply open up Safari and I'm golden.

Another thing I may now be able to live without is a jump drive. If I have an internet connection, I can access all of my files. So what's the use?

Responsiveness / Speed
The solution is a little bit sluggish, but acceptable. To help it out, I optimized my 'local' machine to run as streamlined as possible. A lot of this had to do with the UI. The machine looks a bit retro, buts its tremendously more responsive locally. This should help out.

Windows-to-Windows seems to render fine. On the Mac, you get some dithering: seems to be running at only 256 colors. I'll continue to play with this.

In any case, you can definately get work done. It's not bad.

CNN.com had been set to convenience users who had their monitors running at 800x600. In the early days of the web and for the last number of years, this has worked fine. A majority of consumer and corporate desktops were set to this by default. But increasingly, as more and more users have udgraded their hardware (both with higher-powered video cards and better displays), it appears the norm has now become 1024x768.

To confirm my hunch, I visited one of the foremost authorities on web user statistics - the W3Schools.com Browser Statistics Page. Although their stats come with a very clear disclaimer, noteably that 'You cannot - as a web developer - rely only on statistics. Statistics can often be misleading. Global averages may not always be relevant to your web site. Different sites attract different audiences. Some web sites attract professional developers using professional hardware, other sites attract hobbyists using older low spec computers', it still suggests a lot to consider when building and maintaining a site.

Since clients often confuse monitor size with available 'screen real estate', it's nice to have the stats to refer to.

A key point in these statistics, however, is that W3Schools measures and samples only visitors to its own site, and their statistics cannot reliably be extropolated on a global basis. And because their visitors are mainly comprised of web workers and professionals, some of their stats (especially pertaining to Javascript and user-agent type) may be a bit skewed. Either way, it does give you a good glimpse.

One thing also to consider as far as what format and screen resolution to support is there are work arounds. For instance, liquid-based design conforms to nearly any size. Its drawback is the difficulty in maintaining a 'tight design'. Sites utilizing it take longer to construct and the conceptual methophor changes dramatically. It's also tricky to implement due to limitations/differences in browser implementations of standards, especially CSS.

In the future, browser support for the standards should become more consistent. While this should make implementation easier, coming up with workable concepts will still be a challenge.

Another path to mediate screen resolution is a growing trend of dynamic resolution-dependant layouts. This essentially combines three existing technologies: javascript, CSS and the Document Object Model (DOM). Basically, the Javascript samples the user's agent or browser, gets its screen resolution and them adjust the CSS dynamically. All of it happens on page load. It's more work and requires scripting skills and knowlegdge of the DOM.

While I'm very envious - as you are - the point that I wanted to highlight was that of process. According to a story I recently read about it, he started off by selling pixels to friends and family. When he was able to obtain enough money he issued a press release, that according to CNN 'was picked up by the news media, spread around the Internet, and soon advertisers for everything from dating sites to casinos to real estate agents to The Times of London were putting up real cash for pixels, with links to their own sites'

The point is that the idea was simple and original, and used an extremely basic process. There's still hope for us normal folks after all...

In my own little world, I had somewhat of a similar experience at work. No, not with the money, but with creating media buzz. See my entry about Advocacy Through Public Relations to see what happened.
My point is that there seems to be a process. It builds on itself. Kinda like a domino effect. It was a blast.

Now if I can only come up with something to profit with....

I discovered an excellent web magazine covering modern web development in TreehouseMagazine. What really struck me was how they've used basic Acrobat PDF to present their core product - essentially a magazine.

Normally, many newsletters are designed in portrait mode on 8½ x 11. Shifting to landscape mode however, yields huge gains in viewability and usability, especially for publications geared to be read from a website.

Even at 125% magnification, landscape shows nearly all of a standard page. This changes the overall experience for the reader and makes reading more enjoyable.

A couple of factors are driving this. With high resolution monitor settings getting to be the norm (1024x768 and higher), landscape mode performs beautifully even at 125% magnification. Almost 95% of the page is in view at this range. What's more, is that you gain a sense that you're not missing much. This changes the experience of reading the publication. It's more enjoyable to consume.

The only drawback I can find relative to this option is for publications that are cross-purposed. (Meaning the print version is put online for download or vice versa.) Readers might fret over this when a hard copy is generated. Either way, a standard 8x11 fits the bill.

If you are publishing an type of publication that will be distributed or experienced electronically, conisider this format as a very strong option.

Every once in awhile I'm made aware of something I've managed to forget. In this case, it was Acrobat's ability to be viewed at full-screen mode.

Information Week's online rate card presentation offers an excellent example of what PDF can do and how it can be used to jazz-up an online presentation. Plus, with Acrobat's straightforward creation tools that most anyone can master, even sophisticated interactivity is reachable for everyone.

What struck me was how the presentation went into 'Full Screen' mode and you immediately got the impression it was high tech. (At least that's what happened to me.) I click a few pages until I wanted another resource, when then the familiar Acrobat tools kicked in. It was a eureka moment for me.

Simply pressing ESC anywhere in the presentation windowizes everything again. You see it's just Acrobat. An added bonus is the obvious ability to maintain great printability for the user. Magically, the presentation is 8.5 x 11. Imagine that!

Now this is great design. Simple, stable tools used in a creative way. Great job InformationWeek!

When you really think about the following tips, it starts to run deep. Like Zen deep. Some of these are basic; others more involved.
Here's what they boiled down to - my summary:

1) Beware of New Domains - New domains are in Google's sandbox for 9-12 months. This is not BS. It's the perfect time to build content and fine tune content workflows. Realize this fact for new ventures. Yahoo and MSN are much more friendly these days.

2) Feel Your Audience - This helps you write content and copy geared towards them. What they're looking for. Finding you becomes easier.

3) Pay Attention To Info Architecture - How you design and adapt the information and navigational system on your site matters. Try to integrate what people are searching for into your navigational elements/architecture.

4) Pay Attention to Interaction Design - Think of this as a combination of #3 and technical writing. Design your interaction elements (link names, alt tags, link text) with your audience and terms in mind.

5) Make Sure Your Site is Friendly to Spiders - That razzle-dazzle database-driven content probably sucks as far as spidering goes. Human readable URLs, clean directory structures, strong inter-linking are important. Make it easy on the bots...

6) Use Your Title Tags with GREAT Care - These are weighed heavily. Use them with care and creativity. It's not just about jamming them with every keyword and hoping for the best. Description phrases are also important.

7) Publish Good Content That Promotes Links - Everybody has something worthwhile to say. Publish content that offers insight, infor or interest. People will link to you because of it. The more links the better.

8) Don't Be Obsessed with Number 1 - Anywhere on the first page is good. (I actually prefer being 3 or 4). This is where #6 comes in.

The Internet Archive's Wayback Machine brings you back - sometimes as far as 1996 to the dawn of when the World Wide Web grew popular. You can find some interesting pages.

Up until recently, the first version of Grabers was up there. No longer available. The earliest was 1998. Somewhere I have an old hard drive or some zip disks possibly with older versions. The following screenshot shows generally what the site's homepage looked like. This version lasted until mid-2001, when we updated again.

One of the first site's we helped work on back in 1996 was a site for the Progressive Media Group. Scott Herman of Common Grounds Internet Cafe helped up produce it. Below is their homepage (circa 1996)

As I locate additional archival stuff, I attempt to post. I have my original hard drive from Grabers along with an assortment of zip disks. Both are projects for a Sunday afternoon....

Two key points.

1)Gooogle's 64 character page title display limit
2) Finding Synonyms and 'Ignored Words'

The 64 Character limit
In organic SERPs from Google, there is a 64 character limit in what it shows in the page title. After that, elipsis are used (...) Best to keep your titles within this limit. This is a soft limit and page titles can be somewhat longer. Other engines, Yahoo, MSN, etc. use longer limits. The actual structure of the tag may want to have the targeted phrase towards the beginning of the tag. This may be somewhat fluid. Use your own style.

Google Synonyms
Don't know if I spelled that correctly - too busy to look it up - but a powerful tool in Google is the synonym locator. It shows you what Google considers as synonmyms. Use the tilde (~) before the word. SERPs list them in bold. These can help locating words Google considers the same as others.

Faked CSS transparency is accomplished by utilizing a transparent GIF file for the background of nearly any element. This provides you with the ability to acheive translucentcy online.

To do so, however, the respective CSS elements must render the background with each call to the server. (Actually, not all browsers do this - only the dominant IE). This oftentimes cloggs the browser and rendering performance drops.

Typical example includes a 10x10 pix semi-transparent image used as a background on a wrapper DIV. IE renders the background from top to bottom (ala repeat-y) - but a user can see this being 'painted' to the screen at request time. Netscape's and other browsers performance in this regard is markedly better.

The solution to performance issues like this is to respect that IE is truly a dog and to give a file it wants - namely a large onek, instead of a small one to be wallpapered. While this in itself might lead to other performance bottlenecks, it appeears to bne easier work for the browser.

IE's Issues with Cached Backgrounds
By default, IE's 'Temporary Internet Files' settings force the browser to look for new versions of web pages by DEFAULT. Because of this, background images used in pages are not cached by the browser, and are requested anew during each page request. This leads to an issue known as 'CSS Flicker', and can be quite bothersome for navigational elements. Background image reloading also tends to crash the browser if the user hovers over these areas more than just a few times. Other browsers don't recall CSS background images by default -they cache them. Performance - such as :hover - is smooth.

For a better description, http://www.webreference.com/programming/css_flicker/2.html, or better yet - http://www.fivesevensix.com/, or still http://wellstyled.com/css-nopreload-rollovers.html

From this research, this issue can be fixed rather easily on Apache-based servers with little bit of code in a standard .htacces file. Be advised, however, that ExpiresActive is not extremely common within an Apache build, but can be compiled in very quickly and easily. This worked like a charm!

+++++++++++++++++++++++++++++++++++
ExpiresActive On
ExpiresByType image/gif A2592000
ExpiresByType image/jpeg A2592000
ExpiresByType image/png A2592000
+++++++++++++++++++++++++++++++++++

Consider the following - used to as a link to a different section of a page.

‹a href="#anchor"› The is the link text ‹/a ›

Generally, when linking to another part of the page, you insert an 'anchor tag' where you want to jump to. Like so:

‹a name="anchor"›‹/a›

Here's the point: if have a markeup element with the ID set to say 'anchor', there is no need to add the ‹a› ‹/a› tags. Such as

‹h1 id="anchor"›The Title Blah Blab ‹/h1 ›

The navigation still works. No need to add anything else. The ID is the anchor name. A subtle but powerful point.

This is another example how CSS and the DOM change the coding paradigm within well built pages.

Usually, I from the camp of getting it done as quick as possible. In competent web design, however, this can have implications. The mantra of modern web dev is heavily based in semantics - basically how a page is built through its markup.

When working quick, I take proven trails that sometimes need to be re-thought. Especially from a sementics perspective. Anyways, I alwasys used two (2) ULs floated left and padding to be a two-column effect.

The article suggested a much simpler one-list move. It floats the LIs forty-fifty percent of the UL width. Very simple, very effective.

I often overlook simpler solutions in code. This helps to become more aware of the possibilities - check it out:

ul.button{
list-style-type: none;
width: 550px;
margin: 0;
padding: 0;
}

ul.button li {
float: left;
width: 40%;
margin: 10px;
border: 1px #000 solid;
padding: 5px;
text-align: center;
background: #f3f3f3;
}

Domain masking is simple - the URL shown in the address bar is a custom URL - like www.grabers.com - but the content of the page is from another URL - such as www.xyz.com. There's all sorts of reasons why this could be useful.

This has been around for along time. But what is a lesser known fact is that search engines will use the meta information in the search listings somehow. I hadn't been aware of that and this fact is important.

According to HostingDude.com, "But masking is even more powerful than this! You can do more than just hide the domain name to which your customers arrive. You can also control the Meta Tags which are read by search engines, so that your new website can be found, ranked, and cataloged under the keywords and description you desire.You can insert a Meta Title, Meta Description, and Meta Keywords, and these will be passed to any search engine finding your site. This helps create a separate identity for your site from the site to which you're forwarding, and improves your site ranking. "

This is somewhat open to discussion, but nevertheless still relevant as many search engines still rely on the meta information is some way. Jill Whalen at High Rankings gives an excellent perspective on the current state of affairs in this regard

South by Southwest 05 is a series of conferences covering a range of media - including Interactiive technologies - and holds a yearly conference. The upcoming one is in the second week in March and features conferences and talks by a range of people of who know their stuff.

That type of list really distills down a lot of the talent that is curently out there. How far away am I?

Very interesting

The tool is called Nichebot and allow you to find keyword niches that can be used in building traffic on a website. These types of things were once the realm of WordTracker exclusively...it's nice to see that tools as powerful as this can be playee with without much risk.

Generally, adhereing to strict standards-compliance principles when building sites these days takes time and testing. And a lot of it. I'm constantly seeking ways to refine and reduce the effort in getting something to look good quickly.

I've known about sprites for awhile, but never needed to use them. But the site's interface came from an outside deisgner, and I was committed to implementing in using standards.

To deal with the main navigation, I decided to rely on sprites to drive the interactivity. In essence, a sprite is a box - a hollow box of sorts - that is used as a template to show things in. With CSS, simple manuevers allow you to alter the content of what's in the box (its position, for example) in response to user behavior. The box can hold an image that is much larger than its opening.

What used to take either tons of slicing and messy javascript can now be accomplished very cleanly.

Many windows users use Outlook Express on the desktop to send and receive emails. Since I am currently running a somewhat antiquated machine - a Win 98SE box - this might come in handy for users still stuck on that platform.

For Outlook Express 5

Go to: Windows/Application_Data/Indentities/{whatEverNumber}/MicrosoftOutlookExpress/. Thats the folder that houses everything.

A quick investigation on how more modern versions of Outlook operate have yielded the same. Local Settings/Application_Data/.... So it basically the same.

What I do is use webmail. After awhile - when its gets 'full - I download into Outlook. Then it can be archived forever if needed.

I've found myself going through emails from 2002. If I took the time t o hook up my old hard drive, I'd have 3 years worth of email.

For what you ask? Posterity.

I post this in response to today's CNN article on the growing junk email problem, titled Study: Spam costing companies $22 billion a year. Essentially a survey of 1,000 US adults were conducted and the results were startling.

What I found interesting was the fact that a good number of people actually read some spam messages to 'just see what they say'. This is a bad idea. Many of these emails can contain code that allows the sender to snoop on the recipient. Although there's little that can be done programmatically, it does give them an idea on who reading. Because of this, they send the viewer more. Key point: curiosity killed the cat - just delete junk mail.

Another thing that I pointed out was the fact the spammers can send mail into a domain without it needing to be addressed to an actual address. All Grabermedia accounts send this email in the trash automatically, but this can reconfigured. Look into this setting. (Cpanel/email manager/default address).

Email spiders and/or bots that troll the web looking to harvest email addresses are also a huge spam source. Computer program visit sites, harvest email addresses and leave. Those emails are put on a list and you get spam. This can be guarded against by either limiting the number of email addresses published on your site (a bad idea in my opinion) or via an excellent tool I found that encodes the addresses. Using this tool, the robots get confused and harvest nothing.

The excellent encoding tool is provided by Raleigh-based Automatic Labs and is a free service. Basically you can take the address and input it into their form and out comes the code to encode it. You place that code in your webpage. I use this tool a lot.

Basically spam sucks but there are ways to keep it in check. You'll never eliminate it.

Today's NY Times article gave an excellent summary of the growing number of choices in web browsers. Customizeability is prime time. This article should be a must-read for all web users, especially clients.

The article points out, of course, that even IE can be modified to support tabs. But why spend the $40 bucks...

Beyond the high-level information, I thought it was a shame that they forgot to mention that these options only scratch the surface on what's available. Ironically, there is a great Buffalo, NY-based archive of virtually every browser ever. A browser collection at http://browsers.evolt.org

Take your pick.

My commercial sites these days are opting to go with fixed base page widths that snap to the center, instead of liquid design that consume more browser realestate. Grabers uses this technique in the page you are viewing.

Although fluid design certainly has some advantages, implementation is often a less-than-comforting. Designers want to know how much space they got; a 'jello' design allows them to.

Although the entries really didn't provide any competitive insight, it does provide a glimpse on what's on the radar screem with some of the executives.

Then there's Red Hat People Blog. In this one you get a better feel for the company and what's going on under the hood - especially for Linux power users.

It'll be interesting to watch as more organizations use these.

The first site I visted, was essentially an ipodcasting (ipc) aggregator (iPodderx.com). Event podderx goes on to say "Many podcasts are largely aural recreations of conventional weblogs by the bloggers themselves."

Then I found ipodder.org and downloaded a podcasting receiver app. The one I got seems to be having problems...I hope I didn't screw anything up...

Basically weblogs in Audio format..that's what it is. So the next question is how many of the 10M users are savvy enough - or have the inclination - of receiving such content?

This is a good question...interesting

Here it is.

Heavy css use implies the use of a lot of images. These are called through css. Create a directory structure within a styles directory. Doing so will allow the css system to be more self-contained for the webmaster. This might help increase portability. It might speed execution (my conjecture).

Any way, I did it such as:

.categoryMain {
background: #fff url(i/c.jpg) 0px 0px no-repeat;
padding: 15px;
}

But why even use subdirectories?
Continuing on this thought, why even use a subdirectory. Just place the images in same directory as the css file. Now you have a cleaner path for subsequent DOM moves. So now its:

.categoryMain {
background: #fff url(c.jpg) 0px 0px no-repeat;
padding: 15px;
}

Its still portable. You case rename the root directory to whatever later. I've always seemed to bumble my way around, especially calling images with css. This makes it simpler.

I have identified a package that will allow me to do this. Still need to determine how permissions are going to work. This is an important issue.