Rick's b.log - 2010/02/14
It is the 6th of May 2024 You are 3.137.157.45, pleased to meet you!

Rick's b.log - version 2

Let's face it,

• The older version SUCKED.
• That index page calendar grid was an eyesore. And if you thought it horrid to look at, don't contemplate what it was like to generate. By hand. 'cos I never got around to automating it.
• Tweaking all of the page headers, links, etc. Aaargh! It's no wonder long lapses occurred in my reporting. So many things I thought of saying, but managing the files was such a pain I just couldn't be bothered.
• God knows the layout was clumsy. What? A link to the next article and a link to the icky-index? Seriously?
  Excuse me while I hork all over myself.

I may be somewhat constrained by not running this as a MySQL project, but on the other hand it has the potential to be really quite flexible. I'm actually liking the base files approach. At least I can concentrate on functionality that is important, as opposed to nonsense like letting commentards vote each others comments. I mean, really, when's that useful in the real world?
Also, I can steer clear of login access control schemes. Maybe some blogs and forums wish to only have members post. I value your privacy enough that I'll leave it up to you how much information you care to share.

And as always, completely advert free.

SEO types tell me all the money I could make by linked advertising. It could pay for a dedicated "Rick" server. Better computer equipment. Maybe even a short rental of the DER-2 so I can make out with a real Japanese woman<cough>-bot...
Maybe so. But I find adverts tend to pollute sites. Sure, there may be a value in advertising for we don't learn stuff by osmosis - but with the toss-up between useful and annoying...
...let's say I run AdBlock Plus and NoScript on my browser. Advertising? What advertising?

Besides, it is a moral question. When people let sponsored adverts and analytics on their site... that's when you know their heart isn't in it any more. They sold out. They are chasing the money.
My website may be a disorganised mess, but it is a pure on-line representation of me. Who I am. So the rule is pretty obvious - no f***ing adverts!

Seriously. I'd take my site down before I added AdWords.
This isn't a teenager talking about no sex before marriage, a vow everybody knows will be broken. This is something I feel strongly about. You come here to... to what? To learn ARM code? To drool at a decade-old picture of Alyson? [don't deny it, my site log says that's why you're here!] Maybe to read the random gibberish that passes for coherent thought. In other words, that mysterious collection of musings known as bee-dot-log.
You don't come here to have some third party I don't know recommend services it thinks you might be interested in. If you want a guy to fix your toilet, you Google it, or look for the Yellow Pages in your region. You don't look at stupid little insert adverts on a website as you're passing by.

That doesn't mean this site is free from advertising. I'll happily give mention to things I like. But these are my personal recommendations. I think you might be able to trust me slightly more than a machine that's been following you around the web.

Example? You try finding good Japanese noodles. Especially if you'd like something a little different. Oh, sure, there's a hundred types of pasta at your supermarket, but not one single solitary soba noodle. Not a one. As for soba with green tea, hah! Forget it.

• http://www.workshop-isse.fr/acheter-en-ligne/

That's the sort of "advertising" you will see. Me talking about something I have experienced myself. Telling the good and the bad. So more a reportage than an advert.

And this? Today? The birth of the new Rick's b.log!

Notes...

• The commenting should work. I had a bit of a whoopsie in it would call the original comment system. ☺
• Don't bookmark anything - it will only be /blog2 until the code has been tested and a few more of the older entries copied over. Then it will be switched to be /blog [done - blog2 is no longer]
• There is, as yet, nothing before January 2010. Older entries have not yet been converted.
  As such, the calendar has been 'hacked' to stop in January 2010.
  I plan to convert entries backwards, and step back the calendar in sync with this. [done]
• GET requests are used (i.e. with ?diary=20100214 in the URL) because you can bookmark those.
• The format is "?diary=yyyymmdd".
  If this is missing, the script will search for and display the most recent posting.
  If this is a valid page, it will be shown.
  If it is not a valid page, you'll see an error message.
  The exception - if the day is "00", you will see a listing of entries in the specified month.
• If there is a parameter, it must be eight digits long ("yyyymmdd"). Anything else is treated as no parameter.
  There are no other parameters. Don't waste your time trying stuff like "?debug=true", it won't work.
  [other, invalid, parameters are faulted now]
• This code was developed using the EasyPHP pack (with apache server) on an SD card plugged into a Windows XP machine. All the debugging was carefully inserted echo() statements.
  When it worked locally, it was uploaded to the live server.
• The old b.log held pretty much everything within one directory. Can you say messy? Can you?
  b.log2 uses directories for better organisation.
  Don't bother looking for these folders. They contain a redirect which will take you back to the b.log2 index.
• And what you see... all in a Sunday's work... and I don't even think of myself as much of a PHP coder! :-)

Well, that's all. You've read about the new stuff happening to my b.log. Now I'd like to hear from YOU... Do you like the layout? Can I improve things? Tell me what you think, either privately (email address at the top of the page) or in the commentary (comment box at the bottom of the page).

Your comments:

Rob, 15th February 2010, 09:57

Well you could at least have put "Alyson" as a hyperlink to the appropriate page..

Rob, 15th February 2010, 09:59

I must agree about advertising ... I've tried it, might even still have some adsense hanging about in some places, (not that I ever see any adsense adverts) but not doing so for the new site. Never received a penny from an advert or referral yet, so why keep chasing them?

Rob, 15th February 2010, 10:03

Must admit, I did like the "next entry" link... was very handy when catching up. Now I've got to scroll up to the top again to try and find this entry on the calendar, so I can then work out which is the next one.. You reducing your workload in favour of increasing your readers?

Rob O'Donnell, 15th February 2010, 10:11

http://www.heyrick.co.uk/blog2/index.php?diary=2010020X brings up a normal page entitled Rick's b.log - 2010/02/0X and empty content.. No error message like when =201002XX

Rob, 15th February 2010, 10:14

No facility for more than one b.glog entry per day?

Rob, 15th February 2010, 10:21

http://www.heyrick.co.uk/blog2/index.php?diary=99990101 displays an empty page entitled Rick's b.log - 9999/01/01 but the calendar shows December 1969. Hmm.

Rob, 15th February 2010, 10:25

http://www.heyrick.co.uk/blog2/index.php?diary=2010-101 whee. November 2009 calendar displayed. Not validating the month?

Rick, 15th February 2010, 23:55

Wow - November 2009 calendar displayed for 2010-101. Well, I have not yet added range clipping to the date code... but anyway... nice one. :-) On a related subject, year=9999 is probably outside of the range of dates supported by UnixTime. Strange it says 1969, though. Didn't UnixTime start in 1970? You liked the "Read next entry" links? Mmm... I'll have a think about how best to implement them, along with a "back to top" link. Would it help if the date of the entry you're reading is specially highlighted? You can only have one "file" per day. For example, this one is "20100214.html". This means each day has one (file) entry. Of course, under different headings there can be different things talked about.

Mick, 16th February 2010, 00:53

I prefer the new look. I'm sure the ability to bookmark will come in handy too as I can never remember where I've read up to. Feel free to dump a cookie in my PC though to save me the bother. <wink> How about one that simply saves the highest date read so far ie 20100201 would be higher than 20091109. That way, I'd be able to start reading where I left off and skip back without deleting my final read page? Are go on go on go on go on.......