This missive is aimed at you folks just dipping your toes into this web-thing, that is, you think you need a web site, but aren't sure how to plunge in (or how much the whole enterprise is going to cost). Whether you've already decided to hire someone to build a site for you or are weighing your DIY options you need to know a few basic concepts.
I wrote this for someone whose goal was posting photos of their artwork online, so while I might emphasize photo galleries more than you'd care, that's ok-- the basic mechanisms apply.
I'll outline what you need (the parts) for a website, some options for meeting those needs, some examples of what that might finally look like, and finally, how your needs will drive the decisions to the above. I'll conclude by singing a mash-up of Green Day and June Christy tunes.
At no point will I pass the hat or in any other way take up a collection.
1. Parts of a Website
The three pieces one needs for any website are:
- A domain or URL
- Hosting
- Web pages and other site files (HTML, images, etc)
Some definitions and expansions on these concepts:
A domain name is the part after dub-dub-dub (that's "www"). My domain is:
pizzabytheslice.com
You already know a gazillion domains: google.com, npr.org, cbc.ca, marthastewart.com, amazon.com, ad infinitum.
Owning a domain means that you've hired a company like godaddy.com or register.com (known as registrars) to inform the world that you exclusively have the use of a particular domain.
At this point we inevitably experience "the phone book/ telephone company" analogy. It's a good 'un, you'll be humming it for the rest of the week, I'm sure.
A domain name is how we associate a human-friendly name, such as "boingboing.net" or "americanathebeautiful.com", with an address that's meaningful to a computer (an "IP address" is the series of numbers like "204.11.50.136" that computers need in order to find one another).
When you type a website's name (such as pizzabytheslice.com) into the address bar of your browser your computer asks another computer "what the heck is this 'pizzabytheslice.com' thing and where do I find it?". This, called a Domain Name Services (DNS) lookup -- is exactly like you picking up the phone and dialing 411.
Let's explore how we use the phone company's directory assistance service. First, we call info or get the operator (a trusted agent we believe to have the most up-to-date info).
An operator, or more likely a stupidly perky computer voice, asks "what listing, please?"
You say: "casa bianca pizza in eagle rock"
The operator then looks up the numbers he has associated with that business or residence in the city you want using his copy of the trusty phone book.
Well, almost the same thing goes on when you ask for a web site. Your web browser (Internet Explorer, Safari, Firefox, etc) asks another computer that it trusts to have a web domain translated into something meaningful, its IP address.
Say for example (again and again)"pizzabytheslice.com". How did your computer find this web page?
Well, the first thing it did was find the appropriate "phone book". In the telephone example they needed the city, state etc. Computers, too, need some general location info to start their search. The "location" part of a web address is the "dot-com" or "dot-org" at the end of the domain name. In the United States we have a gazillion, sorry, lots, including:
- .com (commercial /business websites)
- .gov (government sites)
- .net ("network" site)
- .org (organization, generally non-profit)
- .edu (educational site, meaning schools & colleges)
These are just different "phone books" or ways of breaking down the ginormous virtual geography of the 'net into more manageable pieces (ginormous means roughly a metric shitload, give or take a smidgeon). Outside the US you'll see these localities:
- .jp (japan)
- .ca (Canada)
- .uk (ummmm, oh, sure United Kingdom, I think someone told me once that this means "England" -- gosh, wish they'd just pick one name)
- .tv (yeah, you thought this was one of ours... nope, it belongs to the tiny, tiny island of Tuvalu. Bill Gross over at Idealab! paid them millions to control the sales of their domains, but it is a country. For fifty-bucks their president will tattoo your name on his forehead, too. Bill'd do it for twenty, but he's that kind of guy)
These are Top Level Domains (TLDs)... you know what? You don't need to know this. Forget that I ever mentioned it. Good. Thanks.
Moving on.
Once the computer knows where the desired website generally lives (using the appropriate phone book based on the last bit of the domain name, the TLD) it then finds the exact listing for your site and returns the numbers your computer wants in order to exchange info with the website.
So owning a domain means paying someone to maintain your entry in the appropriate "phone books". Like I said, various companies do this for different fees, ranging from $0 to $35 for one year. Obviously, there's a catch to the free registration, after all, they do need to perform some work for you, and it's messy work, but we'll touch on what free means and whether it's good or not in just a few minutes. But the service they provide is sending out your master "phone book listing" to the various "switchboard" and "directory assistance operators" and then keeping all of these folks in-synch if any changes are made throughout the year.
Sometimes you need more than just a domain to find a site, like a professor's personal web pages on a college website. For this you'd need the complete URL.
A Uniform Resource Locator (URL) is the sexy way of saying "this is the address or location of a specific file". And, yes, to geeks this is sexy. It's more specific than just a domain. A domain just says a particular computer, a URL tells which file on that computer. Fortunately, they've (the smart folks who thought up the world wide web) provided a mechanism for having a default file returned if someone doesn't specify which file they wanted. Often it's "index.htm". that's just a file. nothing special. the only difference is that frequently people tell some software running on their web server (computer where a web site lives) to send this file as a default if a file isn't specified in the request (URL).
Example, the radio station KCRW's website:
a domain:
kcrw.org
a URL with no file specified (the default page on KCRW's site for the Left, Right, and Center show):
kcrw.org/show/lr
a URL that includes a specific page:
kcrw.org/music/charts/charts.html
There, take a breather. We're well underway and you're doing great. Next: hosting.
I just mentioned a web server. Web pages must reside somewhere. Gotta have a computer for every file to call home. As it happens there's very little that's special about the computers that run web sites, they just have some special software running on them that allows them to find files and talk to other computes. This software is the web server and it's almost always the free version called Apache or Microsoft's Internet Information Server (IIS). Forget the names, they aren't important and I saw wisps of smoke coming out of your ears as you reached overload.
All you need to know is that in order for people to see your website the files need to live somewhere -- on a computer running web server software. Moreover, these computers are (hopefully) running 24 hours a day, 7 days a week -- otherwise people get a nasty error that says your website is "down". This just means that the computer was turned off or broken. So what most of us do is hire a company to maintain a bunch of healthy computers that they leave on, day and night, that will store your web pages.
This is hosting.
Again, it comes in free and pay versions.
hosting is a service provided by a hosting company wherein they provide web servers where you can upload your web site pages and files and other people can access them from their web browsers. Hosting companies provide you a set amount of storage space (on a regular hard drive, again, no magic) and allow for a certain quantity of downloads per month. The last bit is the bandwidth which is how much stuff other people are allowed to download per month (a measure of the signal that actually travels across the wires and fiber optic cables that connect all of the world's computers: this is the net, i.e. computer network, that we mean when we say world wide web).
Web servers run all the time, waiting for someone to request information from them.
Finally, there are the web pages. This is that HTML crud you've heard so much about. It's just like a Word document or any other file you'll see on your personal computer. Typically it has a file extension of HTM or HTML. Not clever.
Web pages are really dumbly, beautifully simple: elegant and cool in this simplicity. It's the text you read with some "marks" around the words that tell your web browser how it should be displayed. You can also include links to other web pages or to pictures. Links to other pages are "hyper" links ("hyper" as in extremely active, allowing you to waste hours and hours at a time with little to show for it... hmmm, while doing this the computer is actually hyper, you're rendered rather sloth-like, odd).
For this reason the marks around the text and the links are together called Hyper Text Markup Language or just HTML (language because the people that write HTML really want credit for being able to "speak" with computers, so we give them this some concession, a minor conceit to make them happy and hopefully get them to stop talking about Deep Space Nine and Lord of the Rings long enough to get some work done).
OK, a recap, because by gosh, I've forgotten what we were talking about.
A web site is a set of three things:
- a location designated by a domain name or URL
- that is hosted on a web server
- and is made up of a bunch of picture, sound and text files that use HTML to organize and link them.
The whole thing works like a phone company switch board, with operators that lookup the numbers associated with the names you know and various extensions routing your call to the correct department or desk.
There. That's a web site. That's also the hard part. The rest, choosing how to get yourself one of these cool toys, is quite easy.
2. the options
With no info to go on I'll plunge in with a few quick suggestions as if you had an overwhelming, burning desire to hear me ramble on even more about geek-stuff. For purposes of discussion I'm going to reference my own site PizzaByTheSlice.com frequently, just to illustrate various concepts and approaches.
Immediately you've got two choices for getting your stuff "up on the web": get your own domain and hosting or signup for a service, such as geocites, where you essentially "piggyback" on an existing site.
Starting with the piggybacking options:
Flickr.com and the dozen or so popular and free photo posting sites offer decent products, giving you an easy way to upload photos onto their site and provide you with a simple means to add titles and descriptions. Most accommodate a customized URL. For example, I have some photos on flickr:
http://www.flickr.com/photos/pizzabytheslice/
As you can see this serves up pictures easily, and allows you to add some text and capture visitor comments. Flickr, owned now by Yahoo!, allows you to host several sizes of your images, too.
The downside to this is that the appearance is set, completely locked; you can't modify the appearance at all.