Twitter allows a short message to be posted in a timely manner. Folk following the tweets, as they seem to be known, can have messages posted to their mobile and receive automated notifications and so forth. I’m toying with the idea of providing a Twitter service to the Jobcentre Plus Database Mirror and I’d invite input from those of you who’ve more experience with this than I, for I’m not on Twitter and I’m a very unenthusiastic Facebook user.
My initial thoughts are a single central user with a series of feeds based on Jobcentre Plus Offices. Existing summarising code could be adapted to produce a posting of the appropriate length with a shortened URL to point to more detail. I’ll probably pilot this with a local, to me, office again.
As always, your input is actively solicited.
We’ve used the EURES web-site to do a ‘back-up’ scrape for the Jobcentre Plus Database Mirror for some time now. The EURES site provides a pan-European access to various government Jobseeker web-sites within the EU. In this regard it is something of a mirror for Jobseekers Direct.
A recent flurry of error conditions has seen me revisit this code and install new, better error handling. Most of the problems we’ve been seeing have been ‘500’ type errors which, if you know your HTTP response codes, mean problems at the server-end. As it is their server there’s little we can do about this. Nevertheless, the code should now be more robust. I’ve also incorporated ‘honest’ User-Agent strings, so that if anybody should be inspecting their logs, they’ll know what we’re up to.
While doing all this we noted the ‘ISOC Code’ field. At one stage I thought this to be a manifestation of the SOC Code, now seen on Jobseekers Direct web-site and dutifully scraped. The numbers don’t match-up, though. It could be that they’re employing a different standard for SOC Codes to the UK, but there’s no information on the EURES web-site about this magic number. If anybody can inform me what this is about and if there’s a mapping to UK SOC codes, then I’d like to hear from you. Until then I’ll continue to put them in the
employer_ref field, which was what I thought they were originally.
There is detail-fetching functionality on the Jobcentre Plus Mirror Database’s web-site and it is used to display vacancy information when given a Jobcentre reference. The function used to only provide data for vacancies which were crudely considered to be ‘live’, that is having a closing date in the future or having been posted recently.
This does have the disadvantage that should a bookmark for a particular vacancy be activated, or a HTML reference made, then there’s the possibility that after a relatively short time that vacancy would not be found.
It has been decided to show vacancies in our historic database rather than just the current live one. This now goes back a couple of years and it would allow us to, for example, note interesting jobs in this and other forums without worrying that they’d turn into some kind of ‘gone missing’ page after a couple of months.
Although we don’t get feedback, in terms of e-mails as such, the logs are monitored. From this we see that one of the more popular Technical Notes is the billing retrieval automation one. The Note documented the automated retrieval of a Telephone Bill from the dominant quasi-monopoly telecommunication supplier here in the UK, BT. This Note has been left, although the code in it no-longer works, as BT keep modifying the web-site. They do so as frequently as author changes his socks, every couple of months or so. When they do, they break the scraper that much of the automated retrieval is based on.
BT are required to provide this billing information before extracting payment through a Direct Debit, an automated variable-amount withdrawal on your bank account. This is a statutory requirement in the UK and BT, in common with a number of other Utilities, have chosen to do this by a mixture of e-mail and a convoluted web-site.
As an aside, traces within the meta-data on all this suggest that BT, or their sub-contractors continue to use IE6, internally at least. All very inspiring.
This WordPress blog has been up for a matter of days. It uses the so-called LAMP stack with a back-end in MySQL and business logic in PHP. Although I have to be modest about these things, it seems to be a tad inefficient compared to other database-backed PHP things on other
ZOIS sites. This prompted a little light Internet searching and it was found that caching was an almost universal recommendation, of which Elliot Black’s seem the best. I’ve therefore installed both native caching and WordPress Super Cache. I need to do this before the world, or at least a whole bunch of search engines, discover these pages and bring things down around its ears.
I’ve been slowly going through the Notes and adding a little more style in the shape of navigation side-bars and so forth. They’ll be consistent with the rest of the site and hopefully easier to read. Along the way I’ve taken the time to re-examine some of the Notes and add some updates. The light of experience and hind-sight are always useful things in this instance. This exercise is now finished and I thought I’d report it here; it was a bit of chore and took some time. The new formatting uses Server Side Includes, something which is relatively ancient, but which I’d not used before. But I’m a sucker for autodidacticism, so off I went.
The formatting choices are generally based on the desire to have a reasonably readable web-site which can be read both on large screen wide displays and on small hand-held devices. Readability suggests a page composition setting the text in columns with side-bars and so forth. The columns would adjust themselves to the width of the display without being too wide in themselves. It seems to be universally held that very wide columns are hard to read, which is why magazines, newspapers and other ‘wide’ print-media resort to columns, and why books are generally narrow and long rather than being wide and short.
I have to confess to being influenced by others on the Internet who espouse such things and are more eloquent in their promotion. I’ll leave you, dear reader, to search these things out. Web-style is a smorgasbord, and not everything is to everybody’s taste.
There’s more information on the style choices made both on the Home and Main ZOIS web-sites, and in this Blog elsewhere.
Necessarily this all depends upon Cascading Style Sheets, width properties of various kinds and of floats. Care has been taken to avoid the obvious pit-falls of conveying semantic information in these instructions. The goal being that most elderly browser may be used to read these pages, and even if it sometimes makes a bit of a bog of the pretty bits, the underlying information remains.
I’ve just turned the DNS over, so the Internet at large should be able to get at these humble pages. We’ll see how we go, but hopefully it’ll augment, in a positive way, the other web-sites that have been around for some time.
I live on Kirkgate in one of the properties which has a cobbled court-yard of sorts in front of it. There is a Horse Chestnut tree set to the front of this area and the same situation is repeated a number of times down the street. We’re at the southern end and the street runs north-south. It is indeed picturesque although the cobbled area toward the lower northern end does become something of a car park. The Horse Chestnuts are pollarded. (more…)
Stag House, Kirkgate, Cockermouth
This is a return for the WordPress-powered
Blog. It has still got the same theme, but the old posts have now disappeared.
This post was originally stuck on the front page, as a kind of greeting, but it’s been replaced by something a bit more succinct.
Although the rest of the
ZOIS sites have News type entries, I think I’ll use this as a central consolidated point for it all. I’ll have to be careful, this sort of Blogging software generally makes drivel writing easy. Not that I write drivel, of course.
ZOIS Blog was private only, and largely concerned On-line Transaction Processing (OLTP). And if anybody else is now reading this, then this new one has now become public, obviously. I’ve made a restart on this by re-posting news type entries that has appeared elsewhere on the main ZOIS and home web-sites, in one form or another. They will be joined by assorted previously unpublished material, and new things as they appear.
The chief distinction with former efforts is that it won’t be confined to just OLTP, or indeed the Jobcentre Plus Database Mirror, but to other things as well. Eclectic things. Things which interest me. Oh Dear, self-opinionated introspection, I’ve gone all Bloggy now too.