Presbyterian College > Academic Web Server > Jon Bell > Transit > Technical notes (Web)
I don't use any special software for Web page design or site
maintenance. I write my HTML code by hand, using either TextEdit or Pico
on a Macintosh, or Pico on the server itself. I keep a complete copy of
the site on my PowerMac G4 at home, which runs an Apache web server so I
can view the site with a browser without having to connect to the
Internet. (I still use a dialup modem connection at home.) After creating
or editing pages on my home copy of the site, I upload them (and the
images) to the server using the scp (secure copy) command at
the Unix command line in the Mac OS X Terminal application. Occasionally
I log onto the server's command line using the ssh (secure
shell) command, to create folders, check log files, or do other server
maintenance tasks.
When I started to make these pages in 1995, I had just installed Web
server software (NCSA httpd) on my college's main Internet
server. At that time, we had only one computer on campus that was
directly connected to the Internet, and also had a graphical Web browser.
It was a second-hand Data General Unix workstation which had a copy of an
early version of NCSA Mosaic (Netscape's predecessor). All other Internet
access was via serial connections, using text-based terminal emulation
software on people's PCs or Macintoshes. For the first year or so after
that, until we ran Ethernet connections to the computer labs and faculty
offices, and started supporting PPP dialup connections, the Web browser of
choice here was the text-based Lynx.
Of course, Lynx cannot display images directly. However, it allows you to select a link that points to an image, and download it to your PC or Macintosh for viewing in whatever image-viewing software you have handy. This led naturally to my basic design pattern: plain text-only pages with all the pictures hidden behind clickable links.
I use very little special code for page layout. I use basic HTML markup for titles, section headings, lists, and an occasional table. Recently I started to use a small amount of CSS (Cascading Style Sheet) code to produce a two-column layout like you see on this page. As I update old pages, I'm converting them to this format.
Even though we all have graphical Web browsers now, I still use this plain-text format, for the following reasons:
It's easy. I'm used to doing it this way, and I don't have to fuss with aligning images or worry about browser compatibility. I do validate my HTML code rigorously using the Web Design Group's online HTML validator.
It's server-friendly. With graphics-intensive page layout, when someone visits a page, the server gets "hit" separately for each individual inline image, as well as for the HTML code. With my pages, when someone visits a page, the server gets hit once for the page itself, and once apiece for only those pictures that the visitor chooses to view. This keeps the disk activity and bandwidth load on the server to a minimum, which was especially important in the early days when we had a single machine that acted as mail server, newsgroup server and Web server for the entire campus, as well as providing compilers for our programming courses!
It's user-friendly. In some ways, at least. You don't have to wait forever for inline graphics, Java applets, etc. to download. On the other hand, I can see that small inline thumbnail images would make it easier for people to choose which pictures to view, and I've thought of trying them on a few pages. But it would mean more work making the thumbnails, a bit more HTML code, more hits on the server, and longer download times.
It's search-engine friendly. Search-engine spiders love text with content (lots of keywords), and hate frames, Javascript, etc. Most of my individual pages rank highly in Google for obvious transit-related phrases such as "San Francisco cable cars" or "Pittsburgh inclines" or "Baltimore light rail." Most of my visitors arrive via Google and other search engines, although a significant number do come via the links that other sites have made to mine over the years, some going back to the early days (1995-1996) when there weren't very many transit-fan sites on the Web. And those links do improve Google rankings... so, thanks to all of you who made those links!
This page was last updated on 18 August 2007.
Presbyterian College > Academic Web Server > Jon Bell > Transit > Technical notes (Web)
This page is © 2007 by Jon Bell (jbell at presby.edu), who is solely responsible for its content.