To FootI am only considering text-based informative pages, free of any form of "Art".
IMHO, your web text pages will be better appreciated if you bear in mind the suggestions on these pages.
Source code shown here may not be the latest version.
It is useful to have a local master copy of one's remote (e.g. Demon TAM) Web site, for editing, searching and checking with local, offline tools.
Starting from or MSIE 4 file://c:/ or MSIE 6 c:/ and bookmarking, or similarly with other browsers, one can check one's pages with the browsers one has before uploading them to the Web site proper. Obviously that should be done - but remember that browsers are on the whole intended to accept almost any HTML and display something; for bad HTML, another browser may display significantly differently. There is a Lynx Viewer service on the Web.
Here's an example of a missed mistake affecting meaning - a checker should detect error in the "DEMO" line :-
| IDEA :- | That's not a good idea! | |
| HTML :- | That's <bnot</b> a good idea! | |
| DEMO :- | That's | - your browser |
| SEEN :- | That's a good idea! | - my browser |
Several types of HTML and CSS checkers are available on the Web, for remote use or for download; see for example the W3 HTML validator page.
Another validator : CSE HTML Validator.
TIDY is a multi-platform standards-based HTML processor; see, for example, Tidy from W3 or SourceForge. - it's downloadable for several platforms. Primarly, it is a layout tool.
I prefer to control layout manually, so I use TIDY frequently in a Command Prompt box as a local checker, within try.bat (below).
:: CHK-TIDY.BAT Runs W3's HTML-tidier in CHECK mode. :: (c) J R Stockton www.merlyn.demon.co.uk >= 2006-01-16 @echo off c:\imports\tidy\tidy -f $$$$$$$$.CKT < %1 > nul mtr -n $$$$$$$$.CKT - "No warnings or errors were found." if errorlevel 1 goto PAGEBAD echo CHK-TIDY : "tidy -f %1" - OK. GOTO FINISH :PAGEBAD echo. >> $$$$$$$$.CKT echo ** Above by "tidy -f %1" ** >> $$$$$$$$.CKT echo. >> $$$$$$$$.CKT c:\imports\tidy\tidy -v >> $$$$$$$$.CKT echo. >> $$$$$$$$.CKT mtr -v:cybk $$$$$$$$.CKT :: mtr (MiniTrue) is used like LIST here. del $$$$$$$$.CKT echo CHK-TIDY : "tidy -f %1" did not show the usual success message. GOTO FINISH :FINISH
There, mtr is MiniTrue, see via PC Links Reference.
TIDY can also do accessibility checks; see TIDY -h .
I have written a Batch file (for Win-32; but easily changed to Dos-16) TRY.BAT - it is in my programs Web directory index - to make certain checks on all *.HTM files in the current directory that are datestamped later than the last time that it was run there.
:: TRY.BAT (c) J R Stockton www.merlyn.demon.co.uk >= 2006-01-16 :: FOR EVERY *.htm FILE IN THE CURRENT DIRECTORY DATED SINCE :: LAST TRY.LOG, CHECK WITH TRY-AUX.BAT :: @echo off :: HUNT TRY.LOG e | COLS 2+19 | STOW LASTUSED > nul NOWMINUS oTRY.LOG vLASTUSED rem > TRY.LOG HUNT *.htm q0 f%LASTUSED% e u %1 %2 %3 %4 %5 %6 %7 %8 %9 "TRY-AUX" set LASTUSED= if errorlevel 2 goto ERROR if errorlevel 1 goto NOFILES goto FINISH :ERROR echo HUNT reported errorlevel 2 or higher pause goto FINISH :NOFILES echo HUNT found no untried files echo. goto FINISH :FINISH
:: TRY-AUX.BAT (c) J R Stockton www.merlyn.demon.co.uk >= 2006-01-16
:: USED BY TRY.BAT
::
@echo off
:: Verify HTML :
call CHK-TIDY %1
:: Trim line ends :
mtr -x -d -b- %1 - "[ \t]+(\r?\n)" = \1
:: Locate common typos for fixing :
mtr -w -c+ -i:c:/homepage/respell.txt %1
:: Find long lines :
mtr -x+ %1 - "^[^ ].{78}.*$"
:: Check P tag balance :
call ALL-PEAS %1
:: Check character set :
EIGHTBIT %1
I used TRY.BAT frequently while editing pages in (Win98 and WinXP); recent errors are easier to fix.
I have written a BP7/D3 DOS-mode program CHEKLINX.EXE - it is in my programs Web directory index - to check locally that the destinations of relative A HREF and IMG SRC links can be found, and that an appropriate NAME exists in the destination page, and to warn of links to such as FILE: and C:. CHEKLINX accepts Long File Names - though 8.3 should be enough. It now also understands BASE, and checks for (some of) the files cited in JavaScript.
It knows about <!-- ... --> comment, and as yet makes no other specific provision for ignoring the contents of <script> ... </script> ; but, with later TIDY, they make J<K an error in script. Scripts may need minor adjustment; for example to use J < K and to avoid unbalanced quotemarks.
That is a 16-bit BP7 EXE; a 32-bit D3 EXE, capable of handling larger sites, should be available.
Page http://www.merlyn.demon.co.uk/bad$page.htm is a short test page, with some sample errors to find; I keep no formal public link to it.
2009-10-15 : I have written an HTML page, for execution from a local machine (i.e. use a copy) to spider and check local links and anchors for an author (e.g. me) who is changing the local master of a Web site. For more, see in the page itself, which is still intermittently being developed.
Check Local Links and Anchors.
It can be difficult to see spelling errors in the HTML source. If the page is not horribly technical, just display it in a browser, select all, copy, and paste into a word processor or a mailer with spelling-check facilities. Remember to correct the original and not the copy.
See also in Mainly About Using E-Mail.
TRY-AUX.BAT by using file respell.txt checked for and allows repair of my own commonest typos, such as ot for or ; therefore, I avoid using them anywhere, including script.
John Savard has a simple Guide to HTML.
By default, all Web pages are indexed at Alta Vista and Google.
Prof. Timo Salmi has a collection of miscellaneous GIF labels at Garbo.
The news:demon.homepages.authoring FAQ may still be somewhere.
A less-known search engine, for Official sites (thanks, Uncle).
Web site "htmlhelp" has a list of links to/towards HTML validators.
The HTML validators which I use or have used include :-
There's a validator at w3.org; and http://validator.w3.org/ is said to be a link checking the page it's on. ?
Nick Kew's on-line Site Valet checks HTML, UK spelling, links, etc.
W3 has a page on Existing Tools.
Phil Bradley once wrote (condensed): Someone wrote 'I read somewhere it was possible to have one's HTML checked. Anyone know anything about this?' There's a variety of places you can go for this. I've got a list.".
I did (1998-07-04) an AltaVista search for "weblint" and found, early in the vast set, a list of Weblint gateways, including ones in Germany, Australia, Canada, Japan, & Birmingham. It seemed very useful; the sites, however, differed.
| HTML Checking : Site Valet | LinkScan - Links, HTML | NetMechanic - Links, HTML
CSE HTML Validator |
| TIDY : Raggett, SourceForge | Dr.Watson | Lynx Viewer |
| WDG | HTML Validator | (Validate Me) |
| Bobby | Access Tester | UWRF |
| W3:- Tools List | W3 HTML validator | W3 CSS validator |
| Cynthia:- | Portal |
JRS: Does anyone have URLs for validated recommendations on page markup for such persons? After Tony Morgan: Try :- http://www.rnib.org.uk/wedo/research/hints.htm http://www.w3.org/TR/WD-WAI-PAGEAUTH#Checklist http://www.w3.org/WAI/GL/WD-WAI-HAG http://www.yuri.org/webable/mp-blnax.html And for validating pages for the visually impaired... http://www.cast.org/bobby/ If your page(s) pass the above, you can display the "Bobby Approved 4" icon.
(The W3 pages are long.)
And maybe news:alt.comp.blind-users ?
APART Consultancy Ltd. : systems and services for the blind user.
RNIB - + links.
See also the DDA.
ATRC Web Accessibility Checker by the Adaptive Technology Resource Center at the University Of Toronto.
Tools, which I have not tried, for converting text to HTML are described by Craig Cockburn and by Terry Newton.
Some possibilities were given in
From: "William F. McCaw" <william@mcsquawk.org.uk> Newsgroups: demon.archives.d,demon.homepages.authoring Subject: Re: [Win3] q-html.exe - Quick and easy Text to HTML converter Date: Mon, 8 Sep 2003 02:17:08 +0000 (UTC) Message-ID: <bjgor4$1n6$1@squawk.demon.co.uk>
Some tools exist which can systematise the layout of a valid HTML page source file - see for example in my PC Links Reference.
A local copy of my JS-GRPHX.HTM will accept JavaScript to write on a canvas element and will generate a corresponding PNG file.
:: This locates some function calls possibly duplicated within a page mtr -o -x+ *.htm ^^(function\s+\w+).* = \1\r | sort | dedupe | find /v " 1"
When a Web page is moved, one can place in the old location a brief page which sends the visitor to the new location : an example is my redirect.htm.
Some authors notify those whose pages link to pages or sites which are moving. In such cases, it would be most helpful if the old link were also given; it can then very easily be searched for in the master copy of a linking site, using local tools such as MiniTrue.
You can use engines such as AltaVista and Google to seek out references to your site; you can then mail requests to update when appropriate. When moving or splitting pages, it is good to leave a runt page at the old location, pointing to the new one(s). When moving to a new host, be sure to remove the old pages.
You can use engines such as and Google and AltaVista to seek out duplicates of your pages; and you can demand removal of unauthorised ones. To help the search, look for unusual combinations of words on your pages. Remember, you often don't want old, uncorrected copies around the Net.
People seem to generate all sorts of difficulties for themselves in setting up their pages. A good way of doing this seems to be to use an over-complex and under-competent HTML editor.
I build my Web site in the DOS directory c:\HOMEPAGE\ using mainly files in that directory with extension HTM, but with some files in the subdirectory c:\HOMEPAGE\PROGRAMS\ and other subdirectories.
I transfer the files by straight WS-FTP to homepages.d.c.u without making any changes, except that (by a WS_FTP setting) the filenames themselves become lower-case on transfer. Links internal to my Web site are lower-case and fully relative (no c:\, no http://, no BASE, ...) - see below.
These links then work within my local browsers, and on the Web site proper (someone please tell me if their browser fails to handle any aspect of these pages!) .
As an example, <A HREF="programs/00index.htm">programs/00index.htm</A> gives programs/00index.htm, from any page within the main directory.
I kept my Netscape Home Page set to c:\NETSCAPE\BOOKMARK.HTM and had bookmarked file:///homepage/index.htm, so that I could readily access the draft Web site with a full browser offline. I do similarly with my current browsers, using a specially-written local home page, linking inter alia to index.htm.
I use MiniTrue (32-bit: mtr; 16-bit: mt), see via PC Links Reference, intensively; mainly to search my page set, but also to automate changes and tests (mtr GUI is no good in my XP, but mt is OK).