Tangential Ramblings

It's only a matter of time before we see Pete Doherty's obituary. (For some bizarre reason, I woke up yesterday with the premonition that the Queen had died, but if she had, it wasn't considered newsworthy.) I suppose that's the case with all of us, but it seems that he's in self-destruct mode. Here's a guide to his 27 years on the planet.

On 8 February, he was convicted for possessing class A drugs, receiving a twelve month community order. Yesterday (less than three weeks later), he was arrested again, this time on suspicion of stealing a car and possessing class A drugs. He almost warrants his own item in the BBC's left-hand navigation.

While I don't doubt his talent, his news appeal is waning, for me at least.

As an aside, one thing that is lacking on the BBC's magnificent offering is an obituaries section.

I played football tonight, in quite chilly conditions. Although the weather shows 19°F (-7°C), tonight's weather warning indicates that an Arctic airmass has hit (which we certainly felt), with windchill temperatures hitting -2°F (-19°C).

I left home in good time, but was late due to battling against the winds coming down the West Side Highway. At times, I was almost stationary, although I got home in record time, with a healthy tailwind.

During the match, I left my cell phone in a bag on the sideline. Afterwards, I checked it for messages, and the user interface was severely slow, each screen taking about a second to display. It was a bit like the old orange PC screens used to be, with the previous screen taking a little while to disappear. Obviously Nokias don't like the cold. It's now back to its usual speed, safe in the warmth of our apartment.

Generally regarded as such; supposed. E.g. the child's putative father.

I've only just got around to reading the complete response by Google to the US Department of Justice's demand for data. While interesting, the data requested certainly doesn't appear to be particularly useful in the government's purported aim to understand whether self-regulation is sufficient to protect minors from unsavoury content.

The introduction can be found here, while the full submission, an interesting 25-page PDF, is here.

Professor Phillip Stark, who I believe is the government's statistician responsible for this initiative, gets an absolute pasting, while the government comes across as being similarly ill-informed in its request.

The government suggests that by gaining access to a week's worth of Google queries, along with the URLs of Google's entire index (along with similar data from other leading search engines), it will be able to test its hypothesis. Google's argument to the contrary is compelling and condescending - it seems that the level of emotion increases as the submission evolves.

It will be interesting to see how this pans out, although I feel that Google's argument would have been even stronger if it had steered clear of the emotion, even if their approach makes for more compelling reading.

There's been lots of hype in the US over the American figure skater Sasha Cohen, who earned a silver medal behind Shizuka Arakawa today. All week, I've wondered why I knew the name, despite being far from an avid follower of the sport, and tonight it finally clicked. Although I made the mistake, she is not to be confused with Sacha Baron Cohen, the talent behind Ali G.

I'd like to see S. B. Cohen interviewing S. Cohen along with the accompanying confusion.

In the press conference following Arsenal's 1-0 win against Real Madrid at the Bernabeu, Arsene Wenger was seen to be smiling for the first time since joining Arsenal in 1996.

In this clip, there is a 30-second clip, beginning at 1:33 and ending around 2:03 during which Wenger's face can certainly be construed as smiling, at times even breaking out into a weak grin. Certainly over the last ten years, Wenger has never been seen smiling in public; many argue that the drought lasts much longer, so much so that his face frowns in steady-state.

The Google in China discussion continues. In addition to Chinese newspapers questioning whether all of the necessary paperwork is in place for it to operate google.cn, they are also questioning Google's policy of telling users which pages are censored.

And Google was also in the news yesterday over its rejection of a demand by the US government for access to a week's worth of search logs, requested to show that voluntary regulation does not work in protecting kids from unsavoury content. In response to the government's assertion that access to a list of search words would help understand user behaviour, Google states that "This statement is so uninformed as to be nonsensical". I have to agree.

Back to China. It's certainly an interesting debate. In some respects the Chinese authorities are adopting a similar policy to that taken by businesses in their application of Websense and the like, although you would hope that the criteria for a site being blocked by Websense are somewhat different from those imposed on google.cn.

This site is blocked by my friend's employers. Websense cites two main reasons for blocking: security (including spyware, phishing, attack prevention etc.) and web filtering (including employee productivity, bandwidth management and legal liability). As far as I'm aware, the site doesn't fall foul of any of the security issues. So it must have been deemed legally questionable, have been hammering the networks or have reduced my mate's productivity significantly, or maybe that of his colleagues. I somehow doubt that any of these are true either.

At the end of the day, both g.cn and Websense can be classified as employing censorship, albeit being driven by different end goals. While there's a way around the latter (change jobs, surf from home), is the freedom of companies to censor what they choose any different from China's censorship?

My friend Alan emailed me the other day informing me that the US government spends the average amount of tax generated by one person in 2.4 seconds. The only citing of this I can find is from a site entitled gullible.info, although I can quite easily believe its truth.

So, assuming the figure is taken as an average of everyone (approximately 300m people), that means that the government would take 720m seconds to spend everyone's tax, or 12m minutes; 200,000 hours; 8,333 days or 31.5 years.

Although the maths here is somewhat crude, it's worrying given that the life expectancy is more than double that figure. No wonder the deficit keeps mounting.

I don't have the equivalent stats for the UK, although I expect it paints a similarly bleak picture, although probably not quite as pronounced.

By all accounts the literally condition continues (see earlier post), with Robin Cousins reporting from Torino last night on the Bulgarian ice dancing pair being, quite literally, on fire during their rehearsal. As Steve rightly pointed out, this must be both hazardous and detrimental to good skating conditions.

Finally, again courtesy of Alan, some stats that counter human instinct but that certainly add up. They result in the following baseball scenario being feasible:

- Johnny Damon has a better batting average than Derek Jeter over the first half of the season, and a better average than Jeter in the second half of the season, yet ends the season with a worse batting average than Jeter

The conundrum was given by Microsoft's Ronny Kohavi in a deck about data mining. Here's the proof by example. X/Y =Z, where X = hits, Y = at bats, Z = batting average:

First half of season
- Johnny Damon: 4/10 = 0.400
- Derek Jeter: 35/100 = 0.350

Second half of season
- Johnny Damon: 25/100 = 0.250
- Derek Jeter: 2/10 = 0.200

Total season
- Johnny Damon: 29/110 = 0.264
- Derek Jeter: 37/110 = 0.336

It's an odd one to get your head around, and relies on injuries early on for Damon, and later on for Jeter.

I've always been quite interested in typography, but one subtlety I've never researched is the array of dashes available, nor their correct grammatical uses. This Wikipedia article gives a very detailed explanation, but here's a shorter version.

Essentially, there are three types:

- the hyphen or minus (-)
- the en dash (–)
- the em dash (—)

(Note that the hyphen and en dash may look similar under small font settings, but they're actually different symbols.) The hyphen or minus isn't actually a dash at all. It's used to hyphenate words or as a mathematical symbol.

The en dash is slightly longer than the hyphen, being half as long as the font is high. So if you're using twelve point, an en dash is six points in length. (Its name comes from the fact that this is generally the characteristic of the letter 'N'.) Essentially, it's used where there is a connection between two things.

- Date and time ranges: June–July, 1–2pm, 3–5 years old
- Page ranges: pp. 38–55
- New York–London flight
- Where two words shouldn't be hyphenated, but are associated: mother–daughter relationship
- As a hyphen in compound adjectives, where the adjectives don't refer to one another: pre–World War II, anti–New Zealand

Finally, the em dash is twice as long as the en dash. As such, it's as long as the font is high. It's used in the following instances.

- To mark a sudden 'parenthetical' break of thought, either at the end of a sentence (in which case you use one) or mid-sentence (where you use two). Here, it could be thought to be replaceable by a colon or parantheses respectively.
- To mark an open date range (Dan Harrison, 1973—)

In North American and old-British usage, the em dash should never be surrounded by spaces in the first example above. Although not officially sanctioned, modern practice is to instead use a spaced en dash, which I prefer. It gives necessary space to the break, allowing the reader to breathe and mentally separate what is to follow from that which has preceded.

Just as with the degree sign (°), it's a shame that neither of the dashes has made it to the standard keyboard, leaving writers to copy and paste it from somewhere else, or use some difficult-to-remember keyboard shortcut.

Nonetheless, I will strive to use these correctly moving forward.

I got a little ahead of myself, I think. I put together a proof of concept (POC) — with some valuable help from Rob — before working out my business requirements. Not that the proof of concept was wasted, as I got some useful feedback. Let me explain a little more, without giving the game away.

A couple of weeks ago, I came up with what I thought was a good idea for e-government. Now I have had quite a few ideas that I think are good, but that turn out to be flawed on thinking further, or on sharing with others. This idea seemed to pass the first test, but I wanted to create something that could be used to accomplish the second.

So, I created a little something that I could share with a few trusted people, and this I did. I got some useful feedback, much of it positive, which means that I'll now take it to the next level. Now it would be easy to take the POC to the next level, but that wouldn't be the right thing to do. (Actually, it would be quite difficult, as I soon got out of my depth in the PHP world, but you know what I mean.) The right approach is to define my business requirements and then work out what's out there that can meet these.

So, I've started to do this. The result may validate the platform on which the POC sits, or may mean throwing the whole thing away. Either way, the result will give me peace of mind that I'm going the right way, as opposed to using the first piece of software that came to mind.

Writing business requirements is a pain in the ass, but doing so will likely save a lot of pain and heartache in the future. Getting things right upfront is fundamentally important to the success of a project, even it it seems like taking a step back at the time.

Watch this space!

There are certain sentences that you should never have to read, especially on the BBC's website. Today, I read the following sentence in a BBC article entitled Safe Hands Seaman drops partner.

"Athlete Kelly Holmes was eliminated after a skate-off with Bonnie Langford."

There is so much wrong with this, I'm not sure where to start. So I won't.

In the mail yesterday, I received a letter from the Republican National Committee (RNC) asking why I've abandoned the Republican Party, asking "have I given up?" Apparently, their records indicate that they have not yet received my 2006 RNC membership contribution.

I looked up abandon in the dictionary and found the following definition: to withdraw one's support or help from. That clearly suggests that support was there in the first place. I can categorically state that it wasn't.

My contribution "is urgently needed to give President Bush ... the resources [he] needs to run effective campaigns and win". I'm hoping that my imminent lack of contribution will have the opposite effect; let's hope so.

"If the Democrats regain complete control of Congress this year, they will roll back all the gains we have made - and obstruct the remainder of President Bush's bold second term agenda." Hm.

Finally, when I worked in direct marketing, there was always a conversation to be had around what date should appear at the top of the letter. Mailing times were often extended to benefit from TMIs (Tailor Made Incentives offered by the Royal Mail for heavy mailers), and production time was not necessarily fully within your control, so you had to choose a date sufficiently imminent to avoid it being delivered before being 'written', and sufficiently far ahead to come across as being stale. The RNC has avoided this problem quite artistically by dating their letter "Monday Morning".

Soccer was cancelled this weekend due to the 25 inches of snow that fell on Saturday, the biggest snow-storm on record for New York. Shame. I quite liked this photo.

Unfortunately, my friend Steve, who has been responsible for some of the funniest comments on this site (far more amusing than the posts themselves many may argue), has recently had this site blocked by his employers. We are all saddened by this news, as we will no longer be exposed to his comments.

Hopefully the recent decision by the firewall police will be reversed in the near future, as the content of this site is reviewed and deemed acceptable, thus allowing us to once again be exposed to his wit and mindless trivia.

In the meantime, after some initial confusion over the rules, Steve commented by email on my recent post Best Moments in Songs. His suggestions are as follows:

- Howzabout the 'snare-drum-down-a-lift-shaft' bang in the chorus of The Boxer by Simon & Garfunkel (them again)? "Lie-la-lie... BOOSH! Lie-la-lie-lie, lie-la-lie...etc."
- Bernard Butler's twangly quarter-tone tremolo-bar depress in the chorus of Suede's We Are The Pigs? Understated and simple as custard, but makes the song. And I hate Suede. And so should you given Brett's dalliances with Louise Wener. Or was it the Justine from Elastica? Yes, I think it was probably the latter, thinking about it.

I disagree on the Suede argument, although I was blissfully unaware of his shenanigans with the lovely Wener - is he responsible for the subject of Andy's comment? I'm not aware of their entire catalogue, but they do have some of the best intros to songs ever (another post?). Beautiful Ones and Trash both have awesome starts.

As for the two examples, I'm not versed in either of them. Can anyone else comment on their validity?

Four years after UK online went live with a content managed dual-language site, its successor (Directgov) has introduced a Welsh workflow and seemingly replicated its entire site in Welsh.

While this is admirable, legally compliant and no mean feat, its value is somewhat limited. Let's look at the stats.

Statistics published by the Welsh Office in 1995 showed that just under 600,000 people from Wales speak Welsh. (My working assumption is that few people outside Wales speak Welsh, and if they do, they can probably get by, nay thrive, in English. Although I did once visit a Welsh town called Gaiman in Argentina. Very surreal.)

62% of these claim to be fluent: 370,000. I think it's safe to assume that anyone who's less than fluent is fluent in another language, most likely English. The one statistic I can't find is the proportion of the 370,000 fluent Welsh speakers who are not fluent in English, although I did find two sources indicating that the percentage is low:

- Even among the Welsh-speakers, few residents of Wales are monolingual in Welsh (Wikipedia)
- There are almost certainly no monoglot Welsh speakers, at least not over the age of about four or five [...] (Geraint Jones)

These are the people for whom the provision of Welsh content is critical, as opposed to being a nice-to-have. I don't have any data showing how many of the under-five monoglot Welsh-speakers are actively seeking government content.

As an aside, in searching for information on this subject, I reluctantly resorted to the Office of National Statistics' site, possibly the worst in government. In its left-hand navigation, they have the most amusing dropdown ever: I can select Natural and Built or Environment, but not Natural and Built Environment.

Back to the point. While I understand the importance of protecting the Welsh language, surely there is a compelling argument for providing the same content in a range of other languages, languages that are deemed critical, languages for which there are high numbers of monoglots in the UK. Although this is not legally necessary, there must be a compelling social argument for doing this, especially when the technology is there to facilitate it. This technology even underlies Directgov, although it may need some UE tweaking for right-to-left languages.

Just as the service to request a burial at sea (for which there is a government form) hasn't yet been put online, the government needs to evaluate where its pound sterling can make the loudest noise.

I've been watching the progress of Steve Fossett in his record attempt to fly around the world and then a bit over the last few days. In the attempt, setting off from Florida, he aims to fly around the world once and then tag on a trans-Atlantic leg, hoping to land in Kent this evening. Although entirely pointless, I'm quite interested by the attempt.

When he hit America's Pacific coast in San Diego, I even IM-ed my friend Kate in San Diego to see if she could see him. She couldn't. At work yesterday, Sesh quipped that the only people tracking his progress are me and his mum. Maybe he's right.

My friend Mark has set off on his own round-the-world trip at a more leisurely pace. You can read all about it here, with a permanent link in my blog links off to the right.

My wife downloaded a few tracks from iTunes yesterday. While she sang somewhat tunelessly (equipped with headphones) along to one of them while I cooked last night, I could have sworn that it was Terry Wogan's Floral Dance. Thankfully not - instead it was some Fleetwood Mac song. Must find out which one.

Not much has been going on this week, as my mind has instead been occupied with some serious personal stuff that came from left-field. However I have been putting a little thought and effort into a new project which will hopefully be appearing alongside this blog in the near future. Trying to sense-check it with a few people and tidy it up before its 'launch'. The combination of CSS and PHP is driving me crazy!

Yesterday was one of the worst on record. The Stone Roses' I Am The Resurrection helped get me to work this morning.

I came across a service called browsershots yesterday. You give it a url and after a while, it comes back with that page rendered in a range of different browsers. Here's this site in Firefox 1.0.4 on Linux, IE 6.0 on Windows XP and Safari 2.0 on MacOS X. A bunch of other browsers are queued up, and should be appended to the above page once they've been processed.

This doesn't remove the need for browser labs for site developers and testers, but it does give those with limited resources (including myself) comfort in that things look good on other browsers.

Jakob Nielsen is always held up as the stalwart of usability, and singly stands out more than anyone else I can think of in any other field. Yet in my opinion, his own site (useit.com), has some areas that are in need of improvement.

My laptop monitor is 1280 pixels wide, probably about average. Using the default Firefox font size and a full screen window, the first full line of text in the above article contains 142 characters. Most books contain around 60 characters per line, which allows the reader to pick up the next line without dropping a beat. I've always been aware that on my site, the left-hand column is quite wide, and I think I'm pushing the limit at around 110 characters per line using default settings. However, the jump from 110 to 142 makes a huge difference to readability.

Also, the hierarchy of titles at the top of the homepage is a little confusing. The site's information architecture is not immediately obvious from scanning the homepage, with no navigational element standing out to guide me.

And when I resort to searching, the results page is on a different domain and feels like a completely different place. And this is from the man whose top ten mistakes in web design included the phrase: consistency is one of the most powerful usability principles. The screen has been designed specifically for a narrower screen, and although it clearly indicates that the results are from useit.com, the first one I happened upon took me to a completely different site: nngroup.com.

Sort it out, Jakob!

I stumbled upon a site that has very little purpose, but that is somewhat amusing nonetheless. If you give it a piece of text, it uses Babelfish to translate it into and back out of one language after another, arriving at what you'd hope would be a similar piece of text in English ten translations later. First, it goes into and out of French, followed in turn by German, Italian, Portuguese and Spanish.

My previous post about Firefox comes out of the mincer slightly the worse for wear, as you can see below.

IchWAR in the term to write to laudatory excess of of the trees of the trowel of the beginning to the method in the Firefox I with, separated present, neighboring the satisfaction used, English was turned fullfilled. They are rent hardly with my calculation to inform if this Firefox that a process of important recovers to request downloadet and the this mine into moments" to still recover the processes that "installing and in something; it begins.

It is covering much. Exactly if I did not request the new version actively, I admit that they whom the will happens unpropitious nothing and ringrazio to take through me announced it in the progress. Everything is good.

Or no. Not seriously that one the processes to recover is installed, apprehension of Firefox, according to the requested thing and still begun him. My Extensionen and subjects disappear all. It would not wish to together say the excluded function; I demand for will refrán that they disappear to me. I must remember of him to this downloadete of before, downloade of the right he and elasticity to form seriously my options of the client like like in the first place. They are happy relatively new in Firefox, then I that could emfatizar I, of whom which were and with the 10 minuteren the effort, the task that still I also, where it was.

Ampere hour and one another thing: in still Firefox, the visualization that really to me congratula begins (if) this use the beta version (Firefox 1,5 Beta2) and that one that appears that I have, "volunteered, around to community" of the section a ordered test that it becomes;. To the school when the professor of the university around when freiwilliger to request, could swear that you he had that actively to introduce your hand in the sky, excess the end to take with selected. It looks like that, since then that the school that sae the 14 years, the definition modified here - to fill to the condition of the balance he it is the hour.

Humph!

I was going to write a complimentary post about how Firefox was keeping me informed, using pleasing, unobtrusive English. I just came back to my computer to be informed that Firefox had downloaded an important update, and that it was "installing my updates and will re-start in a few moments".

That's fine. Even though I didn't actively request the new version, I'm assuming that nothing untoward is going to happen, and thanks for keeping me posted on progress. All is good.

Or not. Once the updates are installed, I shut down Firefox as requested and re-started it. All of my extensions and themes have disappeared. I don't mean de-activated; I mean disappeared. I have to remember which ones I originally downloaded, re-download them and configure my user options in the same way as I did originally. Fortunately, I'm relatively new to Firefox, so I could remember what they were, and with ten minutes' effort, I think I'm back to where I was.

Oh, and another thing: on re-starting Firefox, it tells me (actually congratulates me!) that I'm using a beta version (Firefox 1.5 Beta 2), and that apparently I've "volunteered to become part of the testing community". At school, when the teacher asked for volunteers, I could have sworn that you had to actively put your hand in the air to get selected. It seems that since leaving school 14 years ago, the definition has changed - volunteering is now the steady state.

Humph!

Earlier, Alan asked whether I'd compared the user interfaces of Amazon's UK and US offerings recently. No more context than that; just the question.

Apart from the fact that the number of tabs across the US site had become too numerous to be tabbed (32 compared to the UK's eleven), I'd not given it any thought or analysis. Until now.

The UK's offering is the equivalent of a nervous first-year student, while the US equivalent is a confident graduate embarking on the world. The difference is palpable.

First to the tabs. The few that remain on .com have been worked on by a designer; those on .co.uk look like they've been put together by a developer asked to "put some tabs at the top". In the US, they have shading, and although the pop-up boxes that appear on roll-over are quite a shock, the tabs themselves are asking to be rolled over. The UK ones are horrible. They don't change when rolled over and I have no desire to click. Some categories are plural (books), some singular (DVD), and the variable size of the tabs to fit their content is despicable.

The buttons below the tabs in the UK are similarly dreadful - offensive even. The capital letters turn me off, while I find buttons combined with "underline-on-roll-over" patronising.

Below the primary navs (can you have two?), the two sites share more in terms of their look and feel. The US font and sizing is more pleasing to the eye (particularly in the left-hand nav.), but the modules that the text occupies have similar design touches. Nonetheless, little things bring the UK down (again).

The bullets in the left-hand nav. aren't in proportion with their accompanying text, which grates. In the US, a grey line appears below each module's title bar, separating it nicely from the text below. No such line exists in the UK, which detracts, ever so slightly.

The American site is confident and doesn't feel the need for a hard sell. Welcome. We know you'll find what you're looking for. Make yourself at home and see you at the check-out. The UK gives off a somewhat different air. For God's sake, buy something from us. Please!

I find the differences odd, given that they're both controlled by the same software, and the US has such a huge say in everything the UK says and does.

In themselves, most of the points above are minor, but getting them right generates a pleasing experience; getting them wrong makes you want to go elsewhere. When there are so many people looking in your shop window, you want to make it look good.

There will be more to come on this, btw.

January was the biggest month on record for this site. From a content perspective, there were 31 new posts, helped along partly by the introduction of the All things Excel category and partly by the Monty Hall problem. Oddly enough, or maybe not, the previous high from a content perspective was January 2004, with 26 posts.

Usage has never been higher, with 51,052 pages being served in January (an average of 1,646 per day) from 12,875 visits (415 per day). The previous high was 42,120 pages served in July 2005. Traffic has increased by 552% since January 2005, when the site served 7,821 pages.

By far my favourite search term used to access the site was "channel four babe minger thumb scan", successfully directing people here seven times. Not the most popular, but certainly my favourite. India has made it on to the top five countries from which the site was accessed, with 1.0% of known, resolved traffic, behind the US (78%), Mexico (7%), UK (2.5%) and Greece (1.8%). The remainder of the top ten comprises Australia, Germany, Holland, Austria and Costa Rica.

Finally, you may have noticed that I've entered the world of Javascript, incorporating a random book of fact and a random book of fiction off to the right each time a page is loaded. Thanks to Sesh for his help in implementing this. Each book is taken from a list of ten that I'm particularly fond of, although as previously posted, I'm still not overly comfortable with the distinction.