Validating email addresses with regular expressions
Yahoo! can’t get it right and they are a huge name, why do you think you can?
Consumer appliances and custom browsers
I’m having the joy of writing webpages targeted at consumer appliances. Unfortunately, the vendors of many of these devices have not seen the light and thus failed to grab one of the open source browser engines on the market or go to Opera. Either they are rolling their own, or buying browsers which have never seen a desktop computer.
Thanks to a bug in the rendering engine of one major vendor with a name that starts with S, I have to have odd spaces when I make text bold.
foo <strong>bar </strong> baz
However, this causes a bug in the browser used by a different vendor (also with a name starting with S) so I’ve ended up with:
[% MACRO sfix BLOCK; IF product.browser.someVendor; %] [% END; END; %]
foo <strong>bar[% sfix %]</strong> baz
Isn’t that just lovely?
Oh, if only I could just target Webkit and Presto!
The Personal Touch
There are job agencies that treat you like a person, and there are job agencies that…
Flash continues to suck (possibly)
A Flash developer on the subject of touchscreens:
The only potential “solutions” to the mouseover problem are terrible ones:
A) The best case: every Flash app on every site is re-thought by its designers and re-coded by its programmers (if they’re even still available), just for touchscreens.
Really? Are you seriously telling me that every Flash application out there is entirely dependent on being able to point the mouse at things without clicking?
Even if you ignore touch screens, that still violates WCAG 2.0 2.1 by being inaccessible to keyboards.
I have trouble believing that every single Flash developer is that bad.
When SEO fails
I hate most SEO, I really do. Companies are spending too much effort trying to game their position in search results and not enough trying to provide a good user experience.
Take, for example, British Airways.
The use case: I want to find out how much BA will charge to get me to Pisa this summer.
So, I visit a search engine, type in British Airways and skim down the results. Surprisingly, BA doesn’t appear to be number one, or two. The third hit is Google’s standard link off to Google News, then we hit the BA recruitment site.
What is going on? Why aren’t BA showing up?
The answer? They are, and at number one. I’m just overlooking them because they look like spam from a crappy price comparison website.
This is the problem:
<title>Book Flights, Hotels, Holidays, Car Rental with British Airways - BA.com</title>
I’m looking for BA, but BA are so concerned with having good search engine positioning for “Flights”, “Hotels” and other keywords that I wouldn’t associate with BA in the first place, that they pump up their title with those keywords instead of making it a useful title. The only bit of actual title (British Airways) is stuffed almost at the very end, and I’ve given up reading by that point.
Do they really want browser tabs to be labeled “Book Flights, Hotels, Holidays…”? Do they really want the default bookmark label to be “Book Flights, Hotels, Holidays…”? Probably not, but they’ve put so much focus on their search engine position that they’ve made the sacrifice.
Even being at that position in the search results isn’t entirely good for them. I actively sought them out, initially overlooked them, and when I dig through their spiced ham camouflage I found myself wanting to see what their competitors had to offer.
HSBC don't want me to be their customer
The British are more likely to get divorced than switch their bank, but HSBC are doing a fantastic job of driving me away.
At the start of December I endured the annual round of “There is a possibility your card might have been cloned, we’re going to cancel it and send you a new one — which will take 5–10 working days.”
This was irritating the first time (especially as I had just reached that stage where I had memorised every detail of the card and no longer needed to pull it out to make online transactions).
The second time was just as bad.
This year, they decided to do it just as I was about to start my Christmas shopping.
That was the point where I applied for a new credit card, with a different bank (it was approved today).
Being the end of the year, I thought I had better order myself a season ticket for my rail travel for next year (before the fares are hiked by the rail companies). So, I duly fill out all the forms (including the fun of working out how I’m supposed to enter the photocard id code — hint: The spaces that are printed on the card aren’t in the code you have to type in).
Remember, this is an order for a physical ticket that will be posted to the address where my credit card is registered.
Then I get to the payment page where I enter:
At that point, my transaction is declined.
Did I misremember the Masterphish password? What has gone wrong?
My phone starts buzzing, but diverts to voicemail before I remember that I left it on the other side of the room.
Time to hit the “Call last number” button:
You were called by HSBC, no further action is required by yourself, and there is no need to return the call.
That isn’t very helpful.
I dig out the number for card services and call them back anyway. After a lot of rigmarole, where I inevitably wanted the last menu option, and wasn’t allowed to hear the first menu until I had listed to my balance report, I got to speak to a human … who had to transfer me to another department which made me say “Yes, I did spent that money” for every transaction I made in the last two days (quite a lot, I’m preparing for a party) before believing that I really wanted to spend a large sum on money on an annual train ticket.
Then I got to deal with my railway company’s order system again (naturally, I had to start from scratch).
As I wrote this, I received a voice mail (sadly, there is significant lag on the O2 voicemail system):
Good afternoon, my name is something incomprehensible in an accent from another continent and I am calling from the bank HSBC and have a prerecorded message for you.
Oh joy, the human touch. I feel so loved.
The HSBC theme tune
Sorry. I’m not thinking of HSBC as a superhero right now. This theme tune serves only to annoy. Especially as it reminds me of earlier this month when I spent ten minutes listening to it on a loop.
Please urgently call HSBC on …
Excuse me? “Urgently call”? How does that square with “there is no need to return the call”?
HSBC have been my bank for a decade and a half. I have a credit card from a different bank in the post. Next year I shall look around for a new current account.
Screen readers and top posting Vs Mutt
Many years ago, my primary email client was a little console tool called mutt. It is small, powerful and has lots of nice features.
One of these features is the ability to identify quoted sections of an email.
It just takes one keypress, (capital) S by default, to skip to the next section of fresh material.
It just takes one keypress, (capital) T by default, to toggle all the quoted material into invisibility.
This is a tiny piece of free software that has had this feature for over half a decade (which is when I discovered the feature).
If Mutt can do it, why can’t screen readers and other email software manage it?
I don’t raise the issue of RFC1855 on accessibility mailing lists. The top posting that is the norm on those technical forums might make it relatively hard for me to follow a thread, and it might be a nightmare when trying to read emails with a dozen quoted signatures and disclaimers over a slow GPRS mobile Internet connection, but I know what happens.
People who depend on screen readers have problems with top posting. Why is they software so shoddy? (And can anybody who uses such software please complain to their vendors!)
CMSs that are too much work
Did I mention that my CMS system was too much effort? Yes, I did, and the evidence is in — I wrote a blog post a month ago and it still isn’t on my main website. I’ll have to run an update tonight (which will probably involve manually fixing a parse error I have with some data I’m importing).
Three Levels of Expression
Russell said in “In praise of fragments” that “blogging is about momentum and ‘more considered’ makes momentum harder” and I’m with him on that.
I have a system, albeit one suffering from some technical issues (my production CMS was designed to generation static files and never got a decent admin interface) which I hope to resolve soon (I have Catalyst running on my production server, and I have the software almost written to take over a couple of bits of my website).
Tweets are transitory, and short. I have ideas that can’t fit into 140 characters, and I feel guilty if I spill over into multiple tweets.
So, to blogging. Blogging needs a front end, it gets squished into Atom, and squished into webpages and squished into content’s pages. I’ve using Tumblr as an interface between me & my blog. I am going to replace it with some Catalyst (removing the need to boot the laptop which currently hosts my CMS and tell it to suck down from Tumblr, rebuild and push out — the legacy of moving between systems and not investing the time in cleaning up the mess sooner).
Where was I? Oh yes. A front end makes it easy to bash some content out, to let ideas flow, and to stop holding on to thoughts until you can find the time to express them the way you really want.
Then we come to the polish. I want to be able to express some ideas with polish. I want to include example pages. I don’t want the publication date to be considered highly important. I don’t want the content to drift down into the depths just because it was written a long time ago. My documents about basic CSS centring and inheritance account for over a third of the hits to my website by themselves.
Perhaps Twitter is my in and out trays, blogging is my file cabinet and I’m left wanting a bookcase to show off the things I really care about.
And perhaps, just perhaps, I should take a leaf out of “The death of the boring blog post?” and stop being quite so consistent with how I format that stuff which I care about.
As the new CMS takes over, I think I may be looking at alternative ways to handle the articles section of my site.
New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.
Keep up with the latest Advogato features by reading the Advogato status blog.
If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!