A general grab-bag of tips for using LaTeX, from the days when I still used it a lot:

• To add nice headers and footers to each page use the fancyhdr package.
• Use glosstex for most acronym and glossary needs.
• Use the hyphenat package if you need monospace text (such as that set by the \texttt command) to wrap correctly, with hyphenation marks.
• The varioref package provides easy-to-use cross referencing with automatic phrases such as ‘on the following page…’ etc.
• To include EPS easily in both latex and pdflatex documents, use the following magic incantation in the document preamble:

\newif\ifpdf
\ifx\pdfoutput\undefined
    \pdffalse
\else
    \pdfoutput=1
    \pdftrue
\fi
 
\ifpdf
   \usepackage{hyperref}
   \usepackage[pdftex]{graphicx}
   \DeclareGraphicsExtensions{.pdf,.png,.jpg}
\else
   \usepackage{graphicx}
   \DeclareGraphicsExtensions{.eps}
\fi

Then refer to graphics files, when including them, without an extension. Use epstopdf to convert all EPS files to PDF for inclusion in the pdflatex version of the document.

• Use the custom-bib package to easily create custom BiBTeX style files.
• The following command produces a new environment which gives a nice way of having an introductory paragraph for each new chapter:

\newenvironment{intro}{\sffamily}
{\vspace*{2ex minus 1.5ex}}

As I mentioned recently, I’ve been using Dopplr and TripIt a lot recently as I’ve been travelling more. Although TripIt is far more featureful, I know more people on Dopplr and so keeping it up-to-date is useful as it increases the chance of serendipitous coincidences. They are competitors, but I like them both.

So far it’s been a pain to enter information into Dopplr manually, but they announced today that it can now watch iCal calendars and create trips accordingly. I’ve added my iCal feed from TripIt and it seems to work well so far. They are minor niggles. For example, it gets confused if you have multiple destinations in your trip as TripIt doesn’t really allow for this concept. Also, not all trips involve advance booking, so I don’t bother adding them to TripIt – they’ll need to be entered into Dopplr manually. Nevertheless, it should cut down on the amount of data entry.

As I’ve starting travelling a lot recently, I’ve been making reasonably heavy use of both Dopplr and the less well-known TripIt. The idea behind Dopplr is simple: tell it which cities you’re travelling to and it will share that information with your other Dopplr contacts, notifying you when you’re in the same place. You can also syndicate your travel plans – I have mine published on Facebook and available as a feed via Google Calendar.

TripIt, although it has similar facilities for maintaining a list of travelling contacts, originates from a slightly different and more ambitious idea. Essentially, you email TripIt confirmation emails for hotels, airlines, car rental, etc. – and it parses them and automatically organises them into trips with information-rich itineraries, including weather, maps, city guides, etc. You can print those out, but (more usefully) you can again syndicate them into tools such as Google Calendar via iCal. This way, I end up with details about all my flights, hotels, and so on in my calendar automatically.

TripIt’s not perfect – I’ve found a few bugs – and TripIt doesn’t support every single travel agent (for example, it doesn’t support the one we use within IBM – at least not directly). But the TripIt team are very responsive to feedback – I notified them about a Hertz reservation email this afternoon that wasn’t recognised. They’ve already fixed the bug and the information has appeared in my TripIt account. I’m not sure how they plan to monetise their service (although that’s still not clear with Dopplr either), but I am likely to stay a regular user for the foreseeable future.

(Honourable mention for another useful travel website goes to Kayak, which has the most flexible and useful flight search interface I’ve found).

A long discussion with plv the other day about open source and what it really meant got me thinking about that model when applied to other domains, such as mapping.

Google have clearly made a success of Google Maps (I’ve discussed Google Maps before as compared to Multimap – not entirely favourably – but whatever I think, the market loves the former). Plenty of competitors have also sprung up, notably from Microsoft. Incidentally, Flash Earth brings together all of these services into one ultra-slick interface; although I’d still love to see them available on Jeff Han’s touch screen (iPhone, eat your heart out – your interface is nothing on this).

However, one thing all these services have in common is that the mapping data is (as far as I can tell) commercially licensed, ultimately from a governmental institution. In the UK, we have the Ordnance Survey (who actually produce excellent paper maps, even if their customer-facing technology is a little backward). The Ordnance Survey gets its revenue from licensing data, selling maps, and so on, rather than from general taxation (which is something that as a libertarian I can almost approve of; although it does raise the question of why the government needs to be involved at all, since there’s therefore clearly a market for the data). The closest equivalent in the US appears to be the USGS (which also has other functions).

It always used to be conventional economic wisdom that mapping (or, to be more precise, surveying) was a function that had to be performed by government, because it was so astronomically expensive – in other words, it cost more than the direct revenues one could possibly obtain (presumably the indirect benefit to society is supposedly significant, which is why we engaged in it). Whether you agree with the morality of this depends on your political views, but it is at least plausible. It’s interesting to see that the Ordnance Survey no longer seem to operate on this model, but clearly many folk still believe surveying should be done centrally.

Now technology might be able to change all of this. OpenStreetMap is showing how it might be done – using cheap GPS receivers, driving along streets, and plotting the resultant data (yes, I know the receivers rely on expensive satellites; but there are only a few of them; and they’d be there anyway). Obviously there’s a long way to go, as shown by the short list of places that have been mapped. There are obviously also concerns over completeness, accuracy, and so on (although most of these have an analogy in Wikipedia, too). However, the potential for these maps is huge if the concept does take off – Google Maps mashups would have nothing on the potential richness of data available. The real concern so far has to be over how many people are really interested in creating this data and keeping it up to date.

As with all futurology (aka: guesswork), time will tell.

Update 2006-01-16: A recent edition of the BBC radio programme In Business (available as a podcast) took a rather quaint look at open-source. Worth a listen as a discussion of how hard open-source is to sell, although not as a rigorous discussion of the technological and legal issues.

I’ve recently taken to reading a lot more on-line – particularly as services such as del.icio.us have helped me to find high-quality content and more high-quality blogs come on the scene. This, of course, is the long tail of written content. One of the things I’ve noticed, though, is that as I read more and different things, I get more impatient with long articles. I hardly read non-fiction books any more, and fiction books almost never (preferring film).

I suspect I’m not the only one suffering from this decreased attention span, but the question is – is there anything we should do about it? Insofar as lots of shorter information diverts people from a few bits of longer information (reading 100/articles/week, say, rather than 2 books/week), it probably indicates that we simply don’t get as much value or entertainment from the longer stuff as we thought we did (or should). This is called revealed preference – what you prefer is shown by your actions, not by your words. So I suspect the simple answer is no.

Seth Godin certainly seems to agree with part of this theory – he has a theory that books, in many cases, have now become a ‘takeaway’ for shorter essays and other written pieces. I don’t think it’s fair to go as far as to say that they are simply fluff, but Seth nevertheless makes a good point – that many books simply expand on shorter ideas – and it is questionable, sometimes, what the marginal value of that is over consuming something completely different (everything you do has a time-driven opportunity cost).

The problem, of course, is that building up habits like this may make it harder to concentrate for sustained periods of time on reading/viewing/listening when that is necessary.

I’m interested in what your experiences are – do you suffer from decreased attention span? Is it a result of increased volumes of information, or do you think it’s something different?

I wrote recently about my indecision surrounding the domain of information design; should detail or simplicity win out? (as always, the answer is probably somewhere in the middle – but hey, that’s boring). Google Maps and Multimap provide an interesting example of what I’m talking about.

Google’s maps are simple; straightforward; and link together yellow pages data with mapping data – together with some cool APIs that enable rip-offs (an ancient term for a mashup). However, they also strip out contextual data that any professional navigator would consider important – landmarks, land type, buildings, etc. It seems apparent that Google are trying to hit a particular market – those who find conventional maps too confusing. And to be fair, they’re doing pretty well – empirical data suggests lots of folks use them. Their reasonably robust mobile maps are also quite handy.

Multimap, on the other hand, looks backward by modern standards: a confusing array of mostly-irrelevant fluff crowds the page. They have one strong advantage, however – they provide contextual information, since they simply use OS maps (at least at medium zoom levels). OS maps are excellent for providing detailed navigation data, and Multimap seems to be the only free online provider of them.

I actually use both sites – Google when I’m searching for things (e.g. dry cleaners in Winchester, of which there turn out not to be many), and Multimap when I want a printable map to navigate to somewhere specific. I would hazard a guess that Google’s map interface was designed by someone who works in a city, as they are virtually useless outside one. Google should be able to make the additional mapping data optional as an overlay – after all, there is already a hybrid interface that mixes maps with satellite pictures. This would be a welcome improvement. Multimap should focus on cleaning up their interface if they don’t want to be run out of town.

I’m undecided on information design. For a long time, I’ve been a big fan of Edward Tufte: his wordy-but-worthy book, The Visual Display of Quantitative Information, is a classic of non-fiction (”Best 100 books of the 20th century.” – Amazon), and is self-exemplifying to a fault: the typography is beautiful, the illustrations rich and detailed. It’s also a classic treatise on how to visually abuse statistics. His other publications, whilst they cover some of the same topics, aren’t quite so easy to follow, although the short essay The Cognitive Style of Powerpoint is worth a read if you’re a Powerpoint hater like me (also see Peter Norvig’s Gettysburg Address in Powerpoint).

But Tufte seems only to be liked by elitists (a cynical and slightly cruel view of his work might say that it involves making graphics more complicated). Another hero of mine, Seth Godin, says: Edward and I disagree. He thinks people are a lot smarter than I do. I suspect what Seth is alluding to is the density of data. Tufte loves data that’s dense and rich, and when I’m feeling intelligent, I do too (I love the huge map of London on my wall; I dislike Harry Beck’s classic map of the London Underground for the same reason: why move the stations around to fit a neat pattern? You just lose their position for no good reason: try something like this instead). However, Seth’s approach does seem to be more realistic. Sometimes stuff just does need to be simple: you can’t always have a rich graph with 3 axes and 10,000 data points being a practical solution to a data illustration problem.

Most data presentation isn’t like either of these: it is just bad. Information design isn’t a field that many folks are aware of, but it is important (making plans to invade Iraq by Powerpoint is … well … just inappropriate. It’s hard to know what else to say). It’s particularly important as the world upgrades to Web 2.0, as a large part of the scope for improvement is in information design and presentation. It’s also simply pleasant to see data presented well. I just wish I could decide what the best approach is.

Google have created a powerful brand based on creating simplicity from complexity (what all good IT is about). Their tools aren’t perfect, but they’ve made life easier for billions, and so I think they still deserve some free feedback from time-to-time. So, a few thoughts:

• Mr. Google, please develop a podcast search engine. So much interesting content is now being released as podcasts (quick plug for my favourite: EconTalk), that it would be useful to be able to search them. All you have to do is invent a speech-to-text interpreter that actually works reliably. Simple. [Note: as I sometimes do, I wrote this post in advance of it being published. I've since discovered that such a tool already exists. However, I thought I'd leave the original prose here: Google, if you get one out soon, you could still corner the market]
• Mr. Google, please stop developing so many interfaces – and plug them all together. If I want to do an exhaustive search for something, I now have to search Google Web, Google Images, Google Groups, Google News, Google Video, Google Blog Search, Google Book Search, Google Scholar, and possibly others. This is not a good thing – you’re straying from the simple search you started with. Some of those searches do show up in the main search results, but you could do a better job of tying them together to show what I’m actually looking for. This could be a real competitive edge, especially since the basic searches that MSN and others provide are now actually quite reasonable.

Google still have an edge in providing what people want – for a company so technically-focused, they either have talented marketers or are just lucky. Please, Google, keep it up.

I’ve upgraded my interweb connection to Web 2.0 in the last few months. Although no-one can really point to what Web 2.0 is (even though there’s a validator for it), many people feel that they know it when they see it. I now defend the term against cynics because I think it’s genuinely useful. To me, it’s a combination of little things: blogging and feedreading, a good quality web browser, in-place dynamic web sites (mostly driven by AJAX), to name but a few. For a technically minded person, I’m atypically late adopting, and so I’ve only recently happened on two powerful aspects of Web 2.0:

• Social bookmarking. I now use delicious, so I can manage my bookmarks properly across the many computers I regularly use, as well as discover high-quality sites other people have stumbled upon.
• In-browser Flash-based video. You can argue until the cows come home about the technical superiority of this method, but the success of YouTube shows that it’s the way forward. I use to shy away from web-based video because of the hassles involved: Download or stream? Do I have the right codec? Which media player to use? Flash-based video solves this problem.

However, the Flash-based video technique does highlight a problem that users of Web 2.0 sites are likely to encounter for a few years as they become more prevalent: interface inconsistency. It’s been long recognised that consistent user interfaces are a good thing, but everyone thinks that they can make a better scroll bar, and although Flash has been around for some time, it’s so far mostly been relegated to those websites created by designers who value form over function.

Now that there’s a great use for Flash (no, hang on, really great), and AJAX is becoming more widespread, we will go through some years of pain before the UI conventions are worked out and those who stray are vilified, rather than held up as paragons of originality – it seems to have already started with the hourglass, as well as on some of the cruddier YouTube-imitating sites (several of which don’t let you skip forward in videos). This phase of inconsistency has happened plenty of times before in UI history, although never in an uglier way than some of the first web sites. Brace yourself, it’s coming again.

After euroGel 2006, which was truly a ‘good’ experience for me, Mark Hurst has announced that euroGel is coming back to Copenhagen in 2007. I’ve just booked my ticket, and as a previous attendee, I’ve got a 20% discount, so the price was only USD $480. I’m allowed to share this discount (which is only valid until this Friday, 22nd September) with friends and colleagues, so if you’re interested, let me know and I’ll send you the link. You can sign up at the regular price here.

I’m already looking forward to this conference, and it’s almost a year away.