You are here: Home ->
Technical Stuff ->
Web Technical Stuff
"Web Technical Stuff" is of several types, listed in the table below:
- Web Search Engines
Google™ is slow to add websites or new content. A month after this site was first "published" (i.e., released to the public), the highest rankings were by Yahoo, Alltheweb, Sufwax, Turbo 10, Altavista, Ixquick, and Excite. (All™) To get this site into Google, I had to go through their "upload link" and "upload site map" processes. Then it got listed very quickly. 1
- Site Search Engines
When I wanted to add a site search facility, a couple of friends suggested Google's™. It's easy to add, especially if you know a little HTML, or have an example so you won't be intimidated.
makes it all seem easy, for example. So I added Google search on my computer test site. So here I was, looking at a page with "Hungary" on it, but Google didn't find that page, because it was not yet indexed on the Google internet search platform. So I needed a way to control (or force) indexing this site. (I didn't have a high expectation of forcing Google to do anything.)
So I did an internet search for "website search engine" or something like that, and found 14 before I got tired of looking. Some looked kind of amateurish. Some cost money, and I wanted to see how the free ones worked before shelling out bucks. Some had a 250-page limit, and this website already appeared to have almost 190 pages, excluding side and top navigation bars. (Some of the pages were not in use, so there were actually only about 140 in use.)
Freefind's website looked professional, and it offered daily indexing (although I think I ended up with one every two weeks, plus on demand after adding something I want indexed right away.) The process is essentially the same as Google's™: You register, they send you an account ID and password, you log in, they show you a variety of templates; you pick one, copy and paste the HTML into your HTML editor (I use Dreamweaver™) and modify it if you want. (I put it in a HTML layer so I can adjust the size and location.)
Then you tell the service to index your website, and to not index certain places if that's what you want. (I don't index side navigation bars or PDF files, for example.) It seems pretty easy, thanks to
Freefind offers several versions of sample HTML code to past into your website. I initially chose the version that allows you to choose between a site search and a web search.
The problem with Freefind's "web search" is that it seems to return only sponsored sites; non-sponsored sites may also be returned, but they were too far down the list for me to tolerate, so I removed that option.
Freefind has a wide variety of special features, some of which are listed below. Freefind does updates every so often. For example, I don't remember seeing advanced search when I first installed Freefind Search.
- Advanced search
- What's New lists every page that's changed over the past 3+months, instead of the highlights I provide. (I can't control Freefind's format.)
- Site map in three different formats. I like the table best. But Freefind's map organization doesn't seem accurate to me.
- An index
of all the key words in the website. I wasn't sure what this would be worth, but
it's one way to find spelling errors,
- it's very handy if there are alternative ways of spelling something (e.g., Abramson, Abramsohn),
- if you're having a "senior moment" and can't think of a word, but will remember when you see something like it,
or if you're simply not sure how something is spelled.
Reminder: sidelinks and PDFs are not being indexed by Freefind, because I don't want to exceed the free 250 page limit.
- Other Ways to Search this Site:
Many search engines allow you to search a particular site as follows: search term site:siteURL. Steve Scher, my son-in-law, explained this to me.
For example, you could ask a search engine to search for Scher site:prz.ginsburgs.org. Or use the advanced search feature of many search engines, such as Google's http://www.google.com/advanced_search?hl=en, or Yahoo's http://search.yahoo.com/web/advanced?ei=UTF-8. Google's Advanced Search is easier to find than Yahoo's. MSN's was quite hard to find.1
The problem is that you have no knowledge or control over when and how the site was indexed by the search engine you're using, whereas Freefind will index your site more-or-less on demand.
For comparison, on a particular day, Kronenberg was found on 20 pages using FreeFind, but only 5 pages using Google's site-qualified search, and 7 using Yahoo's.
- Website Hosting: This site is hosted by Startlogic, because the Polonus site was moved there, and Hank Bieniecki, the honcho thereof, was happy with it. So far I am, too.
(But on 2/8/09, Bernard Paull reported that his computer had detected a virus while browsing some portion of this website. Startlogic fixed it within three hours of my reporting it, but it makes me nervous. It was something called the iframe attack, which for other people was a big problem in March of 2008.)
- Text and image size can be made larger by typing "control +", and smaller by "control -" (holding the control key down and at the same time typing the plus or minus signs). In some cases, the images will get fuzzy when enlarged, because they are mostly scanned at 72 dpi, rather than a higher resolution.
Testing for Broken Links.
Dear Diary, today, Sept. 13, 2009, the freefind indexing log indicated there were some broken links. I am embarrassed that I could not remember when I had last checked for broken links — It might have been 20 months ago !!!! This is what happened to me (the anonymous "you" in what follows.)
- First, choose a link checker. There are many out there, but I chose the W3C link checking service, not realizing that it would check only 150 of the nearly 200 pages on this website.
The W3C site has the virtue that you don't have to worry that it might install malware somewhere, or display outrageous amounts of irrelevant ads or provide links to offensive websites.
- The site does not seem to have a decent report that lists only the errors it found and the warning messages it produced. So I had to scroll down the output and laboriously cut and past the info into a spreadsheet for ease of tracking the errors later on.
That was necessary because the tester flagged 76 problems or potential problems on 41 pages, and that's just in the first 150 that it checked !!!
- There are a wide variety of problems this service reports:
- Robot exclusions — some websites exclude robots from surveying the site. For example, I use this feature for sites that are still under development, and that I don't want to publicize.
You have to manually check these, if the link checker flags them.
- Redirections — The link checker will let you know if external links are being redirected, and if the redirection is "permanent" or "temporary." I don't know how it knows the difference. You are encouraged to update your links. I'll probably put the original link in an HTML comment, in case I have to use the original again for some reason.
- Improper Syntax, sometimes flagged as "Illegal Links" or something like that. Fortunately actual web browsers are often tolerant of many common errors. For example, browsers will commonly add the http://www if it is omitted.
- (More to follow next time I do this.)
- To deal with W3C's 150-page limit, I am trying to (invisibly) break this site into three pieces: History and Geography, Identification, and all-else. Checking the summary only box makes it easier to review the results and find the broken links. (24 June 2010)
- Update, December 2011:
- You often have to look at the actual souce code to see what's going on because a) W3C can identify some subtle errors, and b) all link checkers seem to produce some false alarms.
- LinkChecker 7.2, a free program that installs on your computer, can save results in a concise html file, but it also has some weird output. It seems to think some image tags (the ones in withing div tags?) are links without URLs.
- It may not be as complete as the W3C checker, but the free version apparently has a 3000-page limit, which makes it easier to use.
- I'll probably have to use a "mixed strategy" – rotate among several link checkers.
- Named Anchors in Tables: This website has lots of tables for tabular items in the bibliography, images, etc. Named anchors are often used to navigate to a specific entry in a table. Unfortunately, I discovered that sometimes they work, and sometimes they don't. I'm not sure how to fix that.
- Some of the larger scans, especially of the large multiples, are progressive scans, which fill in some of the entire image and then add detail over several more passes. They sometimes take quite a while to fully load, and first appear to be of low quality, gradually coming into better focus and higher resolutions.
- Backlinks are sites which have links to your site (this site, if you'd prefer.) They are explained in Wikipedia.
- If you type backlinks into a web search engine, you'll discover all sorts of sleazy stuff about buying backlinks, how to get more backlinks, etc.
- Find my list of backlinks (still in process).
- Displaying Hebrew Text, 26 Sept. 2010: I added some some Hebrew, דואר, the Hebrew word for mail. Much to my surprise, all my browsers (IE 8, Firebox 3.6, Google Chrome 6.0, Opera 10.62, Safari 5.0) knew to display the Hebrew text from right to left (RTL), but only after I learned they could do that. I.e., if you are entering right-to-left text, you type it in the order you would encounter it if you were reading it that way. I.E. the first letter to type is the right-most letter, but when you type it into a English-language (or other LTR) development tool, you will type it first, i.e., it will show up as the left-most letter when you type it, but will appear as the right-most on the screen.
So דואר is entered
דואר (the dalet is on the left in the source code, but on the right when displayed).
After the site had been public for a year, an msn.com search for Przedborz Stamps still hadn't found www.prz.ginsburgs.org, so I sent MSN the URL on 2 Jan. 2009. This was very effective: when I checked a month later, this website was listed #2 & 3 in an MSN.com search for Przedborz Stamps. But if you merely searched for Przedborz, its ranking dropped to #25 & 32.
Last modified 25 Feb. 2013