A blog on the use of technology at home and at work (and some other stuff)

Mobiles on the tube

I blogged yesterday about how useful it would be to be able to read RSS feeds when out of coverage on the London Underground. I’m hoping that a Google Gears-enabled version of Google Reader Mobile is not far away.

That reminded me of the stories that have been in the press for some time about mobile phone reception being introduced into the London Underground tunnel network. On the one hand, the tube is a haven of tranquility away from the endless interruptions that a mobile phone brings. On the other, there would be a huge upside to being able to use time on the Underground productively – maybe reading news on a mobile browser or answering emails – or being able to get in touch with people above ground when the network grinds to a halt.

So, I fired up Google to see what is happening, and whether or when we might ever have mobile phone reception underground.

BBC News first reported on this topic in November 1999, with the somewhat premature claim that “In a worrying move for many commuters, mobile phones could soon be ringing on the London Underground”, but it was OK because we would be able to vote on whether we wanted it or not.

It all goes quiet for five years, and then, in March 2004, the Register and BBC News report that London Underground is in talks with the four mobile operators.

In March 2005, London Underground publishes a notice in the Supplement of the Official Journal of the European Union (which is where all large government tenders must be published) which states that the organisation is “primarily interested in the possibility of extending mobile coverage to Underground stations, station concourses, passenger walkways, platforms and other ‘off-train’ areas that do not currently have coverage (approx 120 stations), however LUL may, in the future, also consider extending coverage to other areas, such as trains and train tunnels”. They also leave themselves open to offers including high speed data services and technologies such as WiFi, WiMAX, DAB and DVB.

At the same time, Ken Livingstone announces that we’d all be able to use our mobile phones and digital radios on Underground stations by 2008, with a trial at one station due in 2006. I can’t find any further reference to an in-station trial, so it’s not clear if it ever happend.

In May 2005, a TfL press release states that interest in the notice was strong: “over 60 organisations have formally expressed interest in providing mobile phone and new technology services on Tube stations from mid-2008″.

In April 2006, the BBC reports that London Underground is pressing ahead with plans to have mobile phone transmitters in the tube. This is against concerns that they might increase the risk of terrorism (phones sometimes being used to detenate bombs).

In March 2007, a TfL press release announces a trial to take place on the Waterloo and City Line in 2008 and the BBC is reporting that mobile phone coverage would not be extended across the Tube network until at least mid-2009.

Since then, nothing. I wonder if that trial on the Waterloo & City is on its way?

Photo: Nuakin on Flickr. Used under licence.

Reading underground

Nick Piggott blogged last week about low consumer take-up levels of interactivity. He cited examples of “press the red button” opportunities during ad breaks on satellite television and URL calls-to-action in radio adverts.

When people are either enjoying what they were already doing or busy with something else (such as driving the car), then even if they are interested in the product or feature being promoted, they just can’t act on it then.

Nick’s argument is that there needs to be a way to bookmark or tag the opportunity so that you can return to it later.

I travel a lot, and I like my travelling time to be as productive as possible. My laptop is always with me and I answer emails, write reports, and watch downloaded TV programmes whenever I have a few moments spare on a train, plane or boat (I live on the Isle of Wight).

My current commute involves a lot of time on crowded tube trains. A tube train isn’t a great place for a laptop, particularly if it’s standing room only, so the web browser on my mobile phone (an O2 XDA) comes in useful then. I particularly like to keep up-to-date with my RSS subscriptions using Google Reader Mobile.

Sometimes there are posts which include embedded video, a link to a mobile-unfriendly website, or which simply provoke the thought “I must read up on that a bit more”. I tag those stories with ‘Add Star’ in Google Reader so that I can return to them later when I’m back in front of a larger screen.

What we all need, and what Nick is advocating, is ‘Add Star’ for everything else in life.

Incidentally, Google recently announced Google Gears, which allows you to read RSS feeds in Google Reader when you are disconnected. Separately, they have announced Google Gears for mobiles. Frustratingly, they haven’t yet married the two and given us a Google Gears-enabled version of Google Reader Mobile. I hope it’s on the way. A lot of my tube journey is spent underground, and then it’s back to Solitaire.

Photo: Laihiu on Flickr. Used under licence.

The BBC’s website house style

Two months ago, the BBC relaunched their homepage and started the journey to giving their whole website a more consistent look and feel. The homepage has been warmly received, with its widgets, personalisation, Web 2.0 look and feel, accessibility, standards compliance and, of course, a Flash version of the BBC continuity clock of our youth.

At the same time they made the step to move from a left-aligned 800×600-friendly template, to one that was centred and required at least a 1024×768 screen resolution.

The BBC website is a federation of websites, all housed under the bbc.co.uk domain, but managed by an array of internal departments and external producers. It will therefore probably take many years before there is a consistent look and feel across the board, and by then we will be into a whole new web design meme and everyone will have to start again. Indeed, if you know where to look, there are plenty of old sites lurking about under the bbc.co.uk domain.

The people behind the new /programmes section of the website were the next to announce a redesign to bring them into line with the new homepage look and feel.

Next to be tackled was the BBC news website. This was less warmly received. Put simply (and I’m sure there was much more to it than this), they had taken the existing template, stretched it to the new width and added the new standard header and footer. In my view, the result had way too much white space, and didn’t match the designs being rolled out elsewhere. Their argument for a collection of tweaks rather than a complete redesign had been that their users had told them that they liked the current design and didn’t want the BBC to change much.

A number of amendments followed in the days after the redesign, including a reduction in the amount of white space and the restoration of links for weather and sport. It still doesn’t appeal to me, however, not being quite the old site or a new site; I think it is still due a proper redesign.

Other sites on the BBC to have been redesigned in line with the new templates include /help, 1xtra, and the annoyingly-trendily-pointless BBC Three.

On Thursday afternoon, the BBC relaunched its blogs website. The primary purpose was, they state, to upgrade to a new software platform and improve the reliability and integration of the commenting system. What’s really odd is that they have completely missed the opportunity to redesign the section in line with the new look and feel. Maybe it’s a two stage project, with backend first and presentation second, but it seems something of a missed opportunity.

Photo: Robin Hamman on Flickr. Used under licence.

OpenID enabled

OpenID Logo

Like most people, I find having to maintain user names and passwords for dozens, if not hundreds, of different websites a tad tedious. I also suffer the hassle of having one of the commonest-names on the planet. No website ever has davidjones, or something logical like that, available by the time I get to register (with the exception of Twitter – I must have been lucky that day) so mostly I end up with some piece of nonsense like davjon374.

I also loathe websites that insist on pointless restrictions to logon names and passwords (e.g. passwords must be exactly five letters; usernames must contain a number). I don’t mind my banking websites insisting on something specific and complicated – I have something worth protecting there – but I could really do without having to remember dozens of subtely different logons for websites of minimal consequence in my life. I prefer sites that use email addresses as logons, particularly as I went to quite a lot of effort to secure my email domain and I’ve been following the chat around OpenID, which sounds like a very sensible idea.

OpenID works by definining a URI to define your identity. You can get an OpenID from an OpenID Provider. Organisations have formed to specifically issue these (e.g. myOpenID). In addition, a number of existing service providers are OpenID-enabling themselves: for example, I already have an OpenID based on my Flickr account URI (http://www.flickr.com/photos/davidcjones/).

But I don’t really want my very own OpenID to have someone else’s domain name in it, when I have a perfectly good domain name of my own.

So, I’ve OpenID-enabled my davidjones.org domain. I did this using phpMyID and Sam Ruby’s instructions.

I’ve also created my Yadis file.

Twitter, Facebook and Twittersync

Wearing My Twitter Shirt

I’ve been a Facebook addict for some time, and I’ve recently got back into using Twitter. I think my resurgent interest in Twitter has arisen from it breaking a critical mass among my more geeky friends. It also makes me feel popular, as I’ve one from getting one or two text messages a day, to dozens. Yes, mock me if you must…

Twittersync is a cool Facebook application, which allows you to update your Facebook status from Twitter. It’s much better than the official Twitter Facebook application which prepends your status update with ‘is twittering’.

Twittersync lets you filter which of your tweets make it through to Facebook. I have created a little regular expression that looks for tweets that begin with the verbs I use most often in my status updates and only updates my Facebook status with those tweets. It means that other tweets, particularly those of a more micro-blogging style, only appear in Twitter and don’t mess up my Facebook status.

Here’s what my regular expression looks like:

^(?!(is|has|was|likes|would|should|wants|had|thinks)\s)

Don’t forget to tick the ‘is regexp’ checkbox.

Oh, and does anyone know where to get one of those t-shirts? Mine’s a medium, please.

Photo: Niall Kennedy on Flickr. Used under licence.

What lessons from LHR T5 for IT project managers?

Heathrow sign saying 'This facility isn't available right now'

Heathrow’s not-so-shiny Terminal 5 has been in the news a fair bit over the last few days, and for all the wrong reasons.

What lessons can we learn for projects of our own?

1. People are the most important part of any project

Initial reports suggest that the day one issues were less to do with technology and more to do with training, with baggage handlers not being properly trained in how to use their new baggage system. You ignore people in any project at your peril, and yet so many times training and communications are the first things to be slashed from a project plan when budgets get tough.

2. Avoid Big Bang launches if at all possible

Of course BA and BAA will argue that this wasn’t a Big Bang launch, with only 40,000 passengers passing through the terminal in its first day compared to the 82,000 it is designed to cope with. Nonetheless, their biggest trial day had only 2,000 people playing the part of passengers, and it’s a big jump to 40,000.

In my experience of IT projects, Big Bangs are to be avoided at all costs. Wherever possible look for ways to divide the big elements of the project into smaller tasks.

As an example, a few years ago I was involved in a project to upgrade an entire company’s core infrastructure to the latest Windows applications (Active Directory, Exchange 2003, etc.). The largest element of this project was upgrading all of the desktops in the company to new hardware and from Windows 2000 (or earlier) to Windows XP. Our original plan was to combine the upgrade of the desktop hardware and desktop operating system, but we quickly came to realise that it would leave way too much work to perform in the latter stages of the project, it would unecessarily delay some of the benefits to our end customers (faster desktop hardware), and it would hide problems until the last minute.

The one downside was that, on the face of it, it would cost more: two visits to every desk, rather than one. However, looking back at the project I am convinced the approach saved us money, as it meant far fewer implementation issues.

A much simpler example than Heathrow Terminal 5, but with similarities nonetheless. And BA has one or two implementation issues to deal with at the moment. I am sure that they considered moving a much smaller number of routes to T5 on day one. I wonder why they rejected that course of action? I suspect it was to do with cost: it would entail having more ground staff at the airport, with significant numbers of flights leaving from four out of the five terminals. So, they went with a solution that saw them move the majority of their Heathrow operation on one day and, presumably, leave them with lower costs for staffing and passenger transfers.

Of course, savings from less staff training and a larger-scale move will probably prove to be false economies once compensation has been paid to thousands of passengers and significant damage has been done to British Airway’s brand and share price.

3. If it can go wrong, it will

Some of the causes of BA’s problems on Thursday morning were ridiculously trivial for a £4.3bn project: insufficient parking spaces and security posts led to many baggage handlers being late to their posts.

Was anyone maintaining a risk log in this project? If so, were the right people involved in its compilation? Risk logs are often seen as one of the more annoying elements of project management bureacracy but, properly used, they can often be an incredibly valuable tool.

Have a risk log from day one; involve stakeholders from across the business in its compilation; have regular risk log reviews, ideally as dedicated meetings rather than tagged onto an existing meeting’s agenda; make sure every risk has an owner who is empowered to do something about it; learn lessons from previous projects (are you creating Lessons Learned reports?). I don’t know who compiled the risk log for BA’s move to T5 (or even if they had one), but I wonder if they had staff from the front line of baggage handling, car parking and security involved in its compilation?

And, for those events which even the best-managed risk log can’t forsee, have a contigency plan. Was moving flights back to some of the other terminals ever considered? You might argue that the change in terminals is confusing enough for passengers but, given the alternative between being bussed to another terminal, having their flight cancelled, or for their flight to depart without hold baggage, I know which option most passengers would choose.

Photo: Ben Terrett on Flickr. Used under licence.

Organising iTunes

Cat pawing at iPod

Naming conventions are very useful in business IT. Most organisations have a convention for server names and user logons, and anyone who develops much software tends to have conventions for naming variables, fields and functions. Naming conventions can also be useful away from the office. Being in possession of mildy obsessive-compulsive tendencies, I have invested some time in the last few months in tidying up my iTunes library through the employment of robust naming conventions.

When you load a CD into iTunes, it creates a fingerprint for the CD and uses it to identify it in the Gracenote database. This database is a long-standing example of user-generated content (UGC), started quite a long time before UGC became fashionable as part of the Web 2.0 revolution. I think it was Microsoft’s basic CD Player application (from Windows 98) that started the whole thing off.

The challenge is that no-one polices the quality of the data in the Gracenote database, and if you rely on it alone you will end up with something of a mess in your iTunes library. I found myself in this position. It can get really annoying, for example, when you are browsing through the Artists on your iPod and discover seven different ways to spell the same group. Before I could tidy up, I had to set out my own naming convention and now, because I’m a generous type, I’m sharing it with you.

Name: The name of the song as the artist wrote it. If it is a dance mix the name of the mix follows in parentheses. If the song is ‘featuring’ a vocalist, then (featuring Artist Name) follows. I also use the ‘featuring’ option if an album by a solo artist includes some guest duets, as it means that the list of artists is kept tidy on my iPod.

If it is both a dance mix and has a ‘featuring’ vocalist then there are two sets of parentheses. For example: Groovejet (Solar’s Jet Groove Dub Mix) (featuring Sophie Ellis Bextor).

Artist: The Artist’s name, usually as the artist spells it. Wikipedia is helpful for getting this right, as the fans will make certain it’s spelt properly. The important thing is to be consistent in your library, so that when the files are copied across to your iPod the Artist menu doesn’t have loads of differently-spelled versions of the same thing. For classical pieces I tend to include just the name of the orchestra and/or choir; conductors, leaders and soloist go in the notes. It gets to messy otherwise. I try to avoid ampersands unless the artist uses them in their name and I spell ‘vs’ (as in versus) without a capital letter or period.

For soundtracks, I avoid putting in individual artist names for each piece, unless they are famous people in their own right, and instead use The Cast of XXXX as the artist (e.g. The Cast of Les Miserables).

Album Artist: I don’t currently use it. Apparently it fixes the problem where other artists feature on an album and you want to keep the album together under the primary artist, but I find it easier to put them in the title for the moment.

Album: For conventional rock and pop albums this will be the name of the album. I only use this where the whole album exists in my iTunes database. I delete the album name where I just have a few individual songs for a particular artist.

For compilations I usually delete the album name and track numbers; it’s of no interest to me that a particular song came from Now that’s what I call music 46. The one exception is where the compilation has some artistic value of its own, e.g. where the tracks have been mixed by a club DJ, or it is the soundtrack to a film. In this case, I make sure to select all the tracks on a compilation album and set the Compilation option to ‘Yes’. I also set Gapless Album where appropriate (e.g. dance mix CDs and specially produced pop and rock albums).

I do not put things like (Disc 2) after the name of album, I used the Disc Number fields for that purpose.

For classical music, I follow Tunequest’s convention of using the Album field to indicate the name of the piece with the Name field being used to name the individual movements. So, for the Palestrina mass, the Album is Missa Hodie Christus Natus Est and the Name might be Kyrie or Gloria. By doing this I lose the fact that this mass may have been on an album with some additional motets, but I can live with that. The other downside I have found to this approach is that it can mess up the automated album art feature, but that’s not the end of the world as it is possible to tidy that by hand.

Grouping: Supposed to be for grouping movements in a classical piece, but I leave it empty

Year: for rock and pop songs, the year of release. I find these on Wikipedia usually. For classical pieces, the year of the performance or recording.

Track Number: I take care that these are only included where there is an album name and that they are adjusted to take into account where I have ‘faked’ the album for a classical piece.

Disc Number: Only completed where there is more than one disc, otherwise left empty (i.e. I delete 1 of 1 if Gracenote suggests it).

BPM: I’m leaving this blank for the moment.

Composer: Only used for classical pieces. Surname followed by a comma and then the first name, e.g. Palestrina, Giovanni Pierluigi da. By only using this for classical pieces it becomes possible to browse the classical music very easily on my iPod by selecting the ‘Composer’ option on the Music menu. I use Wikipedia to ensure that I spell the composers name the same each time (there can be a lot of variation with early composers where the scholars aren’t consistent on a spelling). I’m not really that bothered by who composed a particular pop song; your mileage may vary.

Comments: Used in classical pieces to indicate conductors, leaders, soloists, e.g. Conductor: Andrew Davis; Organist: Joe Bloggs. Used in dance mix CDs to indicate the DJ that mixed them, e.g. Album mixed by: Tall Paul.

Genre: I try to keep this list relatively small, with most songs going into ‘Pop’. This includes songs that aren’t really pop (and for which Gracenote might suggest Alternative, Electronica or Rock). The difficulty I have is that it all becomes very subjective and classifying them all as ‘Pop’ meets my requirements satisfactorily. My current list is: Choral Music, Christmas, Classical, Dance, Easy Listening, Musical, Opera, Pop and Sountrack.

Album Art: I let iTunes do this once automatically, and have since gone through and tidied up its, sometimes quite random, efforts. I find Google Image search, the online CD stores (particularly Amazon) and discogs.com useful for this (and, in extremis, I can always scan the album art in myself).

Do you have a different convention? Please comment if you do, I’m interested.

Photo: rainspoo on Flickr. Used under licence.

Jones’s First Law

System is down

I studied Chemistry at University. The beer has killed too many brain cells since then, and I think I’d struggle to pass my GCSE. I do recall, however, that scientists were particularly fond of laws, and if you discovered something you got to name it after yourself. Boyle had a law about gases, for example, and Hooke had a law about springs.

Over the years, I have come to observe a number of fundamental truths in the world of IT that I thought I would share with you. Jones’s First Law is a variation on Murphy’s “If it can go wrong, it will”:

IT equipment knows that you are planning to replace it, and will fail a short time before, seemingly just to spite you.

I can detect the bemused looks from many of you. How can a server or switch ‘know’? Surely, it is just a collection of chips, wires and circuit boards. It doesn’t have consciousness or, for the more spiritual among you, a soul.

But my fellow high priests of IT, you know, don’t you? That Exchange server. You had the change requests in place. The business had been warned that they would have to live without email for 12 hours (a dangerous game given the global addiction to Blackberries). So, why did that server, the one ticking timebomb in your server room, the one with no documentation, the poorly-configured storage groups, the flaky backup arrangement… why oh why oh why could it not have held on one more week before giving up the ghost. And taking your weekend with it.

But don’t worry, I have developed a strategy for the avoidance of such unpleasantness. Keep your server in the dark. Never discuss impending doom in its presence. Throw it off the scent by commenting on how reliable its hard disks are, and that they’ve easily got another two years’ life in them. You see, servers are fudamentally vain. They all believe they can live forever, like that NetWare 3 server everyone had, that hadn’t been restarted in three and a half years (oh the heady days before Patch Tuesday). I tell you, it’s worked for me.

I just wonder if I’ve jinxed myself by telling you.

Photo: Jemimus on Flickr. Used under licence.

Hello world!

hello, world

Well, I’ve finally relaunched my blog. I’m not sure why. It seems the trendy thing these days, to have a blog… although, I’m probably about five years too late to this particular bandwagon. Hopefully, this one will last more than the week or two of its predecessor.

Presumably, it is a good idea to set out a purpose, at least in an attempt to stave off the inevitable descent into disconnected ramblings and ravings. Therefore, my blog will focus on my experiences of putting technology to good use in the workplace (I have been an IT manager in media and entertainment businesses for a few years) and also some of my experiences of using technology effectively in my personal life. Occasionally I will permit myself an off-topic meander.

It goes without saying that the opinions I set forth in these digital pages are my own views and not those of my employers, past or present. For the sake of my career, I shall also avoid anything confidential or controversial. “Windows Vista in a corporate environment? How brave!” – that’s probably the level of controversy you should expect.

I’m using WordPress, with a hacked-about version of the Tech Designs 2 theme. And, yes, I’ve nicked the idea of prefixing each blog post with a Creative Commons-licensed photo from someone else.

Photo: Windell H. Oskay, www.evilmadscientist.com on Flickr. Used under licence.

© 2014 David Jones