MainDoctor WhoMusicSoftware
Main Page

Alden Bates' Weblog

Website Management Archives

Page 4 of 5

August 17, 2005

Search Engine Update

You may remember a while back, I posted about how Yahoo hadn't indexed the NZDWFC subdomain very well. I said I'd update with the results in a few months but, um, I forgot.

The results are that the number of pages from the subdomain listed in the Yahoo index has gone from 5 to over a thousand (about 1630, Yahoo reports). This is good. There were only about 2300 redirects last month, as opposed to 4600 in August last year, 432 of last month's resulting from Slurp (Yahoo's web spider) and 140 of them from people searching on Yahoo for stuff. Reducing these values to 0 is impossible, because Yahoo strips the trailing / off URLs, meaning they will always link to (for instance) http://www.tetrap.com/lj instead of http://www.tetrap.com/lj/ and thus causing an unnecessary redirect.

And I see they finally caught up with my music subdomain

Of course, other than the trailing / problem, Yahoo seems to be indexing a lot better than they were last year. And recently announced they had something like 20 billion pages in their index.

OTOH, the www.doctorwho.org.nz domain which points to the NZDWFC site used to show up in Google's NZ index but doesn't any more. I think this is because the domain registration company switched it from doing a permanent redirect to bringing up a page with a meta refresh on it. I have now pointed the domain directly at tetrap.com and I'm doing the redirect myself, so hopefully the NZ domain will reappear in Google's NZ index...

Posted at 8:44 PM | Comments (0)

August 2, 2005

Hosting history of Tetrapyriarbus

I posted this to my LiveJournal at the beginning of 2004 when I switched web hosts, but I thought I'd update it here...

Sometime prior to June 1996 - Planet FreeNZ
I'm not 100% sure about the month. That's the date of the earliest posting of mine I could find on Google Groups which mentioned the URL. During this period I created the Mel Bush page (the first major part of the site), and then TSV Editor Paul Scoones was so impressed with it he asked me to make the NZDWFC page as well.
Oct 1996 - First Doctor Who Web Guide to mention my site. Actually here's the first Web Guide from December 1995!.
URL: http://www.wn.planet.gen.nz/~bates/
Mar 1998 - IHUG
Moved to IHUG and my site moved with me. It expanded a bit during this time, but not a huge amount because it was an ISP page and therefore there wasn't a lot of disk space provided.
URL: http://homepages.ihug.co.nz/~abates/
30 Aug 1998 - tetrap.simplenet.com
This was the first time I bought hosting. Simplenet was a very good web host at the time, though they didn't have Perl, but offered MIVAscript for creating dynamic web pages. It worked quite well. They were recommended to me by Jason Fraser who had his site on there as varos.simplenet.com.
(The simplenet URL first appeared in the 18 September 1998 Doctor Who Web Guide.)
URL: http://tetrap.simplenet.com/
20 Jan 2000 - tetrap.com registered
Due to the impending buying out of Simplenet by Yahoo, as Yahoo were going to scrap the xxxx.simplenet.com subdomains.
Feb 2000 - Simplenet swallowed by Yahoo
Unfortunately from there the service went downhill, with the MIVA server going down, disk space reduced to 100MB, and eventually Yahoo announced they were going to scrap SSI which my site made (and still makes) heavy use of. I therefore opted to jump ship.
20 July 2001 - switched to CIHost
Worst decision I made. Lots of site slowness and unexplained downtime. Unfortunately I signed up for a year in advance. When July 2002 rolled around again, they charged my credit card without asking me. Despite canceling, it took them several months to give me my money back. The only good thing about this period is that CIHost supported both MIVA and perl, so I was able to convert my miva scripts into perl.
6 Aug 2002 - switched to Sectorlink
Sectorlink were pretty good. Anything was a step up after CIHost, but there were niggly little things that annoyed me. Mainly to do with the site statistics system (ISTR this was the monthly reports I was supposed to be emailed not turning up or turning up months late).
6 Jan 2004 - switched to HostForWeb
HostForWeb have been pretty good. Their control panel and stats are pretty good, I can download the raw access logs, and their technical support is prompt. The only thing I think they need to work on (and this has been a problem all along) is that they rarely announce scheduled down-time, thus occasionally I'll find my site is down, report it, and be told that they're doing a server upgrade. Other than that, they're a pretty darn good web host.

I'm working by the assumption that Tetrapyriarbus started in June 1996, so the site is currently nine years old. That's 90 in Internet years. :)

The name, incidentally, was due to the fact that many other Doctor Who sites took their names from planets from the show, usually something like "Frontios" (there were about three web sites named Frontios at the time, ISTR...) and so I picked a planet name from Time and the Rani.

Posted at 10:47 PM | Comments (0)

July 2, 2005

The State of the Tetrap

Most of the traffic to the main part of my site, the www part, seems to be coming at the moment from people looking for LiveJournal icons. I can't think why that would be. :P Also, people really dig the Dalek mood icons. I still prefer the K9 icons, myself. But, like, the average hits to my site keeps going up by the week.

When I started on Host For Web in January '04, my site was using around 1.5GB per month. It's currently eating through 4GB per month... Fortunately my account allows for up to 50GB per month.

My favourite spammers, the Bulgarians switched from comment spam attempts to trackback spam attempts. They didn't notice when I renamed the trackback script and blocked all access attempts to the old one. Apparently they've given up as of the 29th, and I've had very few spam attempts since.

Sadly I think most of the people who googled in to the second item were wanting to build a real one, and #1 is probably people searchign for info on the new series. :P June also saw record traffic to the NZDWFC pages due mostly to the build-up to the Prime screenings, and the exposure this got the site in places like This Week in Doctor Who, which is posted to the newsgroup and several high-profile Who sites. Also the Prime Press Pack went up fairly late in the month, but still managed to get into the top 25 pages with some 143 hits. Himpressive.

Moby and Daft Punk both released new albums recently, and Mike and Enya are both working on new ones. Enigma is... well, who knows? Hopefully his next album won't have the same copy protection that stops me being able to play Voyageur (See also Hotel). Likewise, what Era are up to is not obvious from their Flashturbated site, which is still merrily promoting their 2003 album.

Posted at 3:35 PM | Comments (0)

June 26, 2005

Borked: My Yahoo!

[snapshot of Yahoo! Briefcase]

Along the lines of This is Broken: For as long as I've been using Yahoo! Briefcase, I've had the Briefcase status box on my My Yahoo! page. And for as long as I've had that box there, it has always displayed the same message: "Problem retrieving your personal information. Please try again later."

Clicking on "try again" doesn't do anything, so far as I can see...

Hell, that's not the only thing that's broken on Yahoo! I have the login expiry on Yahoo! mail set to 24 hours. This used to mean after 24 hours, I would have to re-enter my password to get back in. Apparently they have changed this. Now the routine is:

  1. I'm prompted to enter my password, which I do.
  2. I'm prompted to re-enter my user name, password, type in the text from a graphic, tick the "Remember my ID on this computer" (again).

However, at screen 1, I can hit the "sign in as a different user" link, and I'm taken to a screen where I'm prompted to re-enter my user name, password, and tick the "Remember my ID on this computer" - I.E. the second screen above except with no graphic. This doesn't make much sense to me at all - why have a device to stop robots from logging in if it's so easily circumvented? It just makes it harder for me, as a human, to use my Yahoo! mail...

Edit: Right as I was posting this, I went to add a file to my briefcase. There were two buttons at the bottom of the page:

[Cancel][Cancel]

So which one do I press?! (It's the one on the left, but you have to look at the source to find that out...)

Posted at 1:51 PM | Comments (0)

June 16, 2005

When does "next" not mean "next"?

I'm having an odd problem with Movable Type - when I first save a new post, it has a superfluous link to the "Next Post", which actually links to the first post I made. This looks odd, since someone will spot the post on Technorati and click on the next post link, only to end up on my first post (which is now slightly inaccurate, as LJ have implemented categories, as per my previous post, but I digress, often and at length, in fact).

As I say, this happens when I save a new post. If I go edit the post and save without changing anything, the useless "next post" link vanishes. Typically, no one else on the Interweb appears have have had the same trouble.

Must be something odd in the coding of MTEntryNext which causes this...

Posted at 7:58 PM | Comments (0)

May 25, 2005

The importance of checking what you're linking to

There's a page devoted to links to sites of Marvel comics workers, one of whom is named "Mel Bush". I'm fairly certain it's not the same Mel Bush that the Mel Bush page on my site is about, because, for one thing, that Mel Bush is a fictional character and unlikely to be drawing comics for Marvel. The proprietor of that site doesn't appear to have checked and gone ahead and linked anyway.

Not that I'm going to complain about getting links, but if you're going to link to a page, at least check that the page is what you think it is.

For some reason I'm completely unable to load anything on that domain - I presume they've blocked all traffic from Xtra for some reason, because I always get a timeout. Google's cache is a wonderful thing.

Posted at 12:58 AM | Comments (0)

May 17, 2005

The Dead Link Problem

Tonight I wrote a wee program* which scans the tetrap.com site mirror I have sitting on my hard drive, grabs all the external links, and checks said external links by doing a head request. Thusly I can locate and fix any dead links.

Dead links are a big problem on the web. Putting a link on my web page is a risk, because at some point in the future that link has a high probability of going "dead" (the page is gone), redirecting, or even redirected to a porn site. As a surfer, it's annoying to click on a link looking for some information, only to find the page has disappeared, so as a webmaster, I make it a practice of checking my links semi-regularly to make sure they all work and go where intended. Since tetrap.com is pretty large, I have to use a link-checking program...

Which caused me to ponder another question: Why do webmasters have to check their links independently? It seems to me that search engines have already done all the work. They've crawled my pages (so they know which URLs I'm linking to) and they've crawled the pages I'm linking to (because that's what search engines do) so it follows Google, Yahoo!, MSN, etc, already know which of my pages have dead links on them. Wouldn't it be nice if one of them provided some easy way for me to get this information so I don't have to go checking the links myself?

Given the high percentage of pages out there containing dead links, such a service would probably be well received...

* So if you got any funny hits on your web site from "tetrap.com link checka", that was my 1337 programming skillz.

Posted at 1:23 AM | Comments (2)

May 6, 2005

Oh my god

Jonas Luster linked to a CSS cheat sheet. This will make my development sooooo much easier. Also finally I can see how to make paragraphs indent on the first line only, which I'd previously assumed CSS couldn't do, simply because I couldn't find the correct syntax anywhere in the CSS spec.

Now I cry with happiness.

Posted at 12:51 AM | Comments (0)

May 3, 2005

More Accidental SEO

Dear Google,
Thou art on crack.

Google results for Doctor Who Downloads

My crappy downloads page should not outrank the official BBC Doctor Who site. That's... seriously wrong.

Posted at 9:23 PM | Comments (0)

April 28, 2005

Accidental SEO

If you search on Google for Doctor Who fan club, the top result is currently the NZDWFC page. I'm not entirely sure how that happened, but I'm not complaining either. :)

After all, these days Google is everything.

Posted at 9:41 PM | Comments (0)

March 19, 2005

New weblog scheme

OK, I've set up a new bunch of templates for Movable Type, so that my weblog finally looks more like the rest of my site. I've only hit one problem: when you use the search box on the main page, the search results come up in the default MT template.

There isn't a template listed for the search results in MT's configuration, and there isn't any mention of one in the manual entry for templates. Yet in the manual entry for template tags, there's a whole section on using search result tags in your "search templates".

So far as I can tell, you have to construct the search templates as standalone files and upload them yourself. This seems an odd way of doing it when the rest of the templates are stored in the database!

Posted at 5:00 PM | Comments (1)

March 9, 2005

Slashdotted!

My site (or the Mel Bush page at least) was mentioned in a thread about the recent leak of the first new Doctor Who episode. Well, not in a good way, since they were disparaging Bonnie (but that's Slashdot for ya). Grrr!

But Slashdotting ain't what it used to be, so the traffic to my main site merely doubled for a day or so and the Mel Bush page got an extra 200 or so hits.

Also a bunch of message boards linked to the post on the NZDWFC Message board about the leaking of same episode, so a few people have trickled through there... (more, ironically than from Slashdot).

Posted at 7:11 PM | Comments (0)

February 2, 2005

Better hotlink protection

I upgraded the hotlink prevention on tetrap.com to use a method suggested by Smarter Image Hotlinking Prevention on A List Apart. I have a lot of trouble with hotlinkers, mainly from message boards and people linking to album covers from EBay.

The new method allows linking to pictures like so but prevents the use of the img tag for imbedding. I've implemented the script in Perl rather than PHP, and it also checks to make sure the image exists (and returns a 404 if it doesn't), so is a slight improvement over the A List Apart method.

Posted at 8:11 PM | Comments (0)

January 29, 2005

Upgrading Movable Type

I'm scratching my head trying to work out why upgrading Movable Type isn't easier. MT has a website, my MT install is running on a web server which is by definition connected to the web. Why can't I just push a button on my MT install and have it automatically grab the updated files from MT's site?

Or am I missing something? Maybe there is a button and I just haven't seen it.

Posted at 5:25 PM | Comments (0)

January 15, 2005

More Icons

Imagine my surprise after I merged my page of LiveJournal stuff into my main domain (it previously occupied my ISP-given webspace) and got a hold of the site stats to find it is almost the most popular part of my site.

I therefore decided I should probably update it, so I've added a page of Star Trek: Voyager icons (all but one previously posted here), some Greg the Bunny icons to the miscellaneous icons page and another Unreal Tournament 2k4 icon.

Some of them may even be good.

Posted at 12:37 AM | Comments (0)

1 2 3 4 5
Search


Categories

Tetrap.com Site Map