I am available for proofreading existing sites, researching and
writing material, and designing web sites within the limits of my methods.
This site contains thousands of words that I have written over
the years. I have a BS in English with minors in Mechanical
Engineering, Math and Physics. I have a secondary teaching
certificate in English, Math, Physics and Drama (and an MFA in
Theater.) Down through the years I have had several dozen
articles and review published under editorial review and have had
two short plays published that made me some money. From 1977 to
1994 I was self-employed doing small business programming, mostly
in various versions of BASIC but also using assembly languages,
FORTRAN, and COBOL. I worked with companies more in the manner of
a CPA or attorney than as a software company. I was born in 1942.
I can create or review what you have done and provide services
at reasonable fees. I do not accept work I can't get done within an accepted
deadline..
Mike Firth
214-827-7734
Dallas Texas
|
Making a Site Like Mine
- Titles as Sentences
- While file names may have to be short and cryptic, Titles
(inside the HTML) do not. Making them as 40 character
sentences means that the indexing programs, including the
one I use, will produce more readable results. And the
sentence will display at the top of the screen in the
browser.
- Use a good Headline
- A snappy headline will attract the eye and give
information. One of my indexing programs pulls out
headlines and makes a link file. If the first word of the
title is relevant to the topic, it makes it easier to
remember the topic of the site. Making a Site Like Mine
is better than How My Site was Made because Making is
more to the topic than How.
- Design an Outline or Tree
- Know first what chunks you want your site to encompass
then work on filling in the chunks. My site does not show
this outstandingly well, but I do have a vague idea of
what territories are mine and what I want in them.
A site should have depth as well as length - from a
series of base pages, it should be possible to deal with
a series of topics. I dislike sites that spoon feed
information as if the screen does not have a scroll bar
and require that I click a Next Button just to get a
couple of more paragraphs that I have to wait for, which
images could have been loading while I was reading the
first part.
If you provide Prev. & Next buttons, make it clear
where they go. I rarely use a Top Link, since anyone
using it probably can get there faster with Ctrl-Home or
pushing the scroll bar.
- Long Pages Have Jumps
- If you do long pages, as I prefer, then add a Table of
Contents like listing of Links at the top of the topics
in the page. If the page is arranged alphabetically, a
line of links to each letter section is useful.
- Remember the Search Engines
- A lot of people are going to jump into the middle of the
site because search engines can find everything. My site
will come up on almost any search related to glass.
Provide a way on every page to get to the top of the site
(Sitemap in my case) and if Outline or Tree to the
beginning of the branch or major topic. I am not as good
with this. Sitemap is easy, I just wrote a program to
look at every file and if it didn't have a link to
Sitemap, it put one in.
Xeno and other programs will provide a Site outline you
can cut and paste if nothing else works.
- Provide Lots of Hypertext links
- This can be done semi-automatically with assistance from
programming languages or manually. For example, a search
program can return a list of all the words used on the
site for building a glossary and then a program can take
the words in the glossary and go back and turn all of
them into links to the glossary word. This also allows
people who land on a page from a search engine to go
places in your site.
- Use Accessories to Build Indexes
- A number of languages or tools will extract information
that can be easily formatted, if nothing else, dump it
into a Text file and edit it in Word and save it out as
HTML. On my site I index images, headlines, titles and
file names.
- Specify the size of images
- This allows all the text to load while the browser
reserves space for the image to load later. If a size is
not specified, the browser waits till the image is fully
loaded before continuing the text.
- Compromise on Image Quality
- A thumbnail can be so small that it only serves as a link.
I prefer not to use tiny images, but have little need of
huge detailed images, so I reduce the number of colors
and set the resolution at a moderate level 72-100 pixels
per inch. If the size of the image is sharply reduced by
the size chosen above, then I make the image a link to
itself, so clicking on it produces a full sized view in
most browsers. Only if useful, by adding detailed text or
mouse-over, do I consider an HTM file just for showing the larger
image.
- Use the ALT option on images
- This provides a text which will appear above the site of
the picture if it is slow to load, so a user may skip
looking at a slow loading image if desired. It also gives
basic information to blind users and to those who turn
graphics loading off to save time online with a slow
connection. It also is key to building an index for those
who want to jump to images.
- No Frames
- Some browsers do not support frames. Bookmarking a site
using frames is difficult. I think search engines don't
point properly to the right page when they find a
reference, only to the home page without directions. Most
sites using frames use them in a way that could easily
produce the same result with a few lines of header or
footer code added to each file.
|
2001-01-31
It's laborious and far less flexible than
standard page-layout programs in terms of typographic and
other details. I
also dug out a text-editing program I purchased last year,
Adobe PageMill
It is easy to see now how these programs ride on the
underlying code.
Much of what you are getting in PageMill (I think) is what I
am getting in FPExpress and can find in other programs: they are
not "text-editing" programs in the terminology I have
grown up with. (I am, by the way, 58.) Text editing {TE) is plain
Jane shoving letters around. There are people (including me
occasionally) who do HTML this way, looking at the results in a
browser by saving the file and refreshing the browser.
The next step up is word processing. A simple word processing
program "knows" about margins, tabs, indentations, line
lengths, word wraps, etc. Most word processing programs hide the
codes that hold this knowledge although most allow a skilled user
to see where they are (like - I think I recall - Alt F3 Reveal
Codes in Word Perfect) and edit them. When a word processing
program allows "Save As Text" it is doing a conversion
from WP to TE, following some simple rules about ends of lines,
etc.
The next step up is WYSIWYG (What You See Is What You Get -
Wooseewigue) - Seeing on the screen a good or excellent
approximation of what will appear in paper. Now in the Mac world,
this is all you are used to because the Mac was introduced after
dot matrix printers became cheap enough to lay down dots fast
enough to move the image from the screen to the paper. Until
WYSIWYG came on hand, about 1982-83 in the PC world, all WP
involved having codes that influenced a printer, the only way to
see what the printed copy would look like was to print it. To
save this to a file (without the file being huge - a graphics
image of the page) requires complex codes for font, kerning,
size, image location, etc.
But HTML is designed to offer to many different machines the
ability to display a good or better approximation of a well
designed page - what the page looks like depends on who is
viewing it - I may have color or not - I may have a different
choice of font, or not - I may do a text only display - as when I
am blind and having the screen read to me - as my wife does in
fact.
Therefore in HTML, the codes are wide open and (mostly) agreed
upon. There are actually levels of codes - the good old simple
ones, the mildly hot shot newer ones and the cutting edge, drive
at least some people crazy, codes.
It is possible to make a perfectly good and fairly enjoyable page
with about 6 HTML code/command groups - Headline <H1></H1>
thru <H6>, Line break, <BR><P></P>;
Structure <html><body</body></html> and the
href and img src commands.
It is the browser (or editing program) that takes those codes and
images them on the screen - perhaps different for each person. In
the editing program, you can use drag n'drop, click on pointers,
etc., to "edit" the screen and the HTML editing program
makes codes for you as you work.
The power of a WYSIWYG editor is that you can drag and drop and
see instantly what is happening. This was the case when I
couldn't figure out how to get text running down the left side of
an image, when I dragged it until it happened and then went and
looked - the image source command was in the middle of the text
paragraph!
I have one or two questions for when you have time to send
an answer. If I
begin to do a site using PageMill/FPE, is it later possible
to insert other
instructions written in code?
Yes (assuming PageMill makes HTML code) I constantly call up
an existing page, drop into the Source Code viewer to see why
something is done in a particular way, and fine tune or correct
something that I can't seem to do in FPE - for example, get two
images separated that have no space between them, or untangle
some NAME = mess I made. FPE allows entry of unchecked text so
people knowing advanced features can use them (or learn them.)
In particular, I have not yet found in
PageMill anything on how (if at all) it can handle a key-word
database. I
believe you used shareware for this element of your site, so
it seems that a
site might be a combination of elements: some written on a
text-editor, some
inserting other pieces of html coding, some using other
programs (e.g.
shareware); and some, perhaps, even copied and pasted from
codes used for
existing sites. Is that an accurate picture?
Simply yes.
More complexly - the keyword database was built by a search
engine run by a company that is offering me a free service and it
is kept on their machine someplace - their machine comes in and
looks at my ISP's machine and makes up the database. When you do
a search, you get results, I get a service and a list of words
searched for sent to me with an ad and they get ?* (see below) To
add this, I just signed up and cut and pasted the code you find
on the page.
When a person looks at my start page (not the site map) the
counter at the bottom of the page shows a count. This count is
kept by another company that takes a command sent from my page
when loaded, adds 1 and returns the count. I get a small service
and a report on how many accesses have gone on during the week,
you get an suggestion of the popularity of my site and they get ?*
I have no banner ads. They are added by cutting and pasting. They
pay money to the site if a viewer clicks on them. They deliver
customers to the source of the ad. Depending on the ad, the code
to make the ad banner on the screen is a small program - Java -
running on the viewer's machine. This code is typically loaded
from a source shown in the HTML code and run.
Finally, there a programs that run on the ISP web server - I have
none of these because I don't have the right on my server. These
are the programs where you type in a product description (Pliers)
and instead of just a set of links, you get a whole new web page
showing part numbers, costs, delivery time and links to images or
to buy. The HTML code delivered your search word to a "server
side" task handler, which called up the info in a database
search - the results returned were delivered to a formatting
program that made an HTML page that was sent to your machine for
display.
So finally, to have an interactive result, somewhere on the
Internet you have to have the right to run a program to return
results - I pre-format all my stuff off line and up load it.
I have told you about the Mac/Windows thing. I continue to
lean towards
making the dreaded change from Mac to Windows, precisely
because of things
like the shareware issue, and also with a view to
communicating with other
servers when it comes time to upload the site. I know I can
use the Mac
platform for much of this work, but the problem comes in
communication with
others.
I don't know about the limits of shareware on the Mac.
Internet protocols should allow complete freedom in creating
pages and using ftp (I use CuteFTP) via a Mac program to upload
them. Whether this is easy or tough on a Mac, I can't say.
I shall continue to work on planning my site and
practicing with both text
editing and html code over the next week or 10 days, by which
time I will
probably have purchased some hardware to run Windows.
Thanks Mike, for your continuing feedback.
|