Only slightly bent

It’s a well-known – and probably false – meme that the Internet was originally designed to withstand a nuclear attack. But one would’ve thought it could withstand oh, say, an anchor?

Apparently not. This story of severe disruption of the Internet (and television and phone etc) connections in Middle East and Africa is all over the net (the working part) today. Okay, so the anchor-theory is just that – a theory – for now, but it’s plausible. Whatever the cause ends up being, it’s a sad fact that the Internet, which has stealthily grown to become a critical infrastructure that we depend on for a surprising amount of things and yet take for granted, can’t withstand one stupid cable being cut.

One can sort of understand that a water supply is somewhat vulnerable if water from a sewer is pumped to the network directly at the source after cleaning (incredibly, such idiocy happened in the town of Nokia in Finland) since the water network has to be somewhat centralized for source sanitation purposes. But it’s always amazing to see that the world of electronic commerce and most of our modern-world communication hangs, quite literally, by a (fiber-optic) thread.

Many large companies have dedicated “overlay” networks and even guaranteed-bandwidth links connecting their offices. But what good does a VPN tunnel do if it goes – as it would most likely do – over the same underseas cable as the other traffic? Short of Google, I haven’t heard many companies building their own physical backup networks. Maybe it’s time to start thinking about that?

I also wonder when some terrorist group decides that bombs are too much hassle, gets their hands on some bottom-trawling equipment and goes fishing in some choice locations..

Here we go with the food prices

Last year, it was all over the news that food prices are going to go up this year. This self-fulfilling prophecy is now being materialized in a number of locations. In the past couple of weeks alone, I’ve discovered the following few examples:

• Sodexho, our blood-sucking parasite company pretending to serve us food at the office, already “adjusted” food prices last year and has now also hiked the prices of chocolate bars by 10% and soft drinks by 10 cents, which now makes them both about 50% more expensive than at the local supermarket and full 100% (or more) more expensive than at a wholesale place. They have fairly nice profit margins on those it seems.

• Subway hiked the price of their meals and large subs by 40 cents – a whopping 16% increase.

• Stadin Kebab raised the price of their lunch deal from 6.50eur to 7.20eur, another over 10% increase.

Now, considering raw ingredients still make up a tiny percentage of the final costs and hence shouldn’t really affect prices that much, I view these increases as unreasonable. They’re many times more than the general inflation rate.

Does all specialized journalism suck in such a small market?

I can say I’m fairly capable of commenting on the factual contents of news and articles on the ICT and also some other fronts – and I have to say that the general quality of such news and articles in Finland is pretty poor. Too many articles and pieces of news fall into one of the three following categories, relying completely or nearly completely on:

• .. information supplied by one company in e.g. a press release.
• .. one self-proclaimed or presumed “expert”, again with no criticism of the source and without questioning anything they say. Here’s a good example of Taloussanomat quoting Petteri Järvinen on some pretty incomprehensible opinions on broadband services.
• .. on the individual whims and thoughts and ideas of the reporter. Just because you – or I for that matter – think something is or isn’t going to happen doesn’t make it so. Presenting own opinions as industry consensus or a fact is not exactly good journalism.

Even the ones that don’t fall into any of those are often rife with inaccuracies at best and downright lies and false statements at worst. What happened to source criticism? Checking the facts? Researching? Talking to both parties of a debate? Learning your field? The most optimistic scenario that I can come up with is that all those basic rules are out the window when you have to get an article out under too tight deadlines. The more pessimistic / realistic view (take your pick) is that people, whether it’s those writing or those reading, just don’t care about good-quality work anymore.

Or maybe good-quality specialized journalism is just impossible for such a small audience as Finland is. It can’t be simply about sheer global inability to make good specialized articles as in English-speaking markets such things do exist. But in Finland? Extremely rarely.

Where are all the LED-lights?

The compact fluorescent energy-efficient bulbs are slowly making inroads – way too slowly though, thanks to the idiotically limited form factor range that stores here usually sport. However, what I’d love even more than E14-socket spot compact fluorescent bulbs are LED lights. They’d be even more energy efficient than the compact fluorescents and would last longer, too.

So why aren’t there any?! Is it just that manufacturers don’t want to sell you only one bulb in your lifetime? Are there technical issues? Why is it that we don’t have any LED lights for consumers with standard sockets yet?

2½ years is an eternity in smartphones

I had to fiddle an ancient smartphone – the Nokia N90 – for a few minutes and man, did it feel old in so many ways. Extolled for its beautiful display just some two and a half years ago, today it felt like looking at a small postage stamp glued to a brick. I wonder if current state-of-the-art smartphones will feel equally outdated in two or three years?

Speaking of mobile devices, I added my Flickr photostream to the right side over (well, up) there –> on the main page. My Flickr account only gets selected photos taken with my cellphone; some may have short comments, others won’t. Some will be geotagged with GPS coordinates, others will not be. In all cases they’ll be cellphone snaps – photos with real photographic qualities will end up elsewhere.

Why is this important?

With more and more people using the Internet with their mobile phones, more and more sites are launching mobile-optimized versions of their service. The importance and numbers of people using mobiles to browse the web has recently been debated by Tomi Ahonen on his Communities Dominate blog and Dean Bubley who disagrees with Tomi on the numbers (continued here and here).

Best practices checklist

This is all very good, but whatever the numbers are, it’s clear the mobile web is increasing in importance. Personally, after some years of skepticism, already a couple of years ago I found myself using the mobile browser daily. Unfortunately, there are some quite annoying and seriously debilitating trends in that space. If you’re planning on making a mobile-optimized version of your site, consider the following list, which from my point of view captures some of the most important best practices when doing so:

• Make a light-version of your site – light in terms of size, that is. Make images smaller, optimize the page structure, skip some videos, cut some ads etc. Websites exceeding one megabyte per page are not a joy to download over any wireless connections. Yet, don’t make the pages too small either – few things are more annoying than having to read a short article that’s split to something like seven pages to keep things “light”; the RTT delays over wireless networks are way higher than over fixed and page transitions are slow.

• Don’t skimp on the content. Having a “light” site doesn’t mean you should leave out content from it. Some service providers apparently think that mobile users don’t need all that content – wrong! It’s a sure way to annoy people if you put one tenth of your content or shortened versions of articles on the mobile site.

• Take it easy with the ads! I’m all too well aware that mobile advertising is one of the big things in both advertising and mobile operators’ future business. Small, relevant advertisements are acceptable and might even be welcome, but simply porting the current web-situation to the mobile side is unacceptable. I’ve seen some non-intrusive, well-placed and relevant ads on NY Times and some downright horrible ones on other sites. Also, remember that even a relatively small ad takes up a huge portion of the small mobile screen, so keep things small.

• Don’t use Java. I thought Java applets were dying anyway, but it seems at least some braindead banks (ahoy, Sampo!) plan on requiring Java as a “security” feature. Having Java applets perform any critical function will effectively block out all mobile users.

• Consider a client if your service would significantly benefit from it. A case in point is GMail – it works tolerably via the browser, but it works wonderfully with its own downloadable client. Not everything is ready to be run on browser on the mobile side quite yet.

• Make the redirection to the mobile site automatic.. This is one of the most important things and if done right, does not conflict with the point below on being able to access the full site. On the way to making mobile web work well, there understandably have been – and will continue to be – some wrong turns. One of these is the launch of .mobi TLDs (top-level domain), meant for mobile-optimized websites. This is, in my opinion, entirely unnecessary – users shouldn’t be forced to enter a different address when using a phone to access some site. Those running a website shouldn’t be forced to run the same content in two different domains. If you must have the mobile site at a different address without automatic redirection, use simple and short addresses – things like m.site.com are light years faster to type on your phone than, say, http://www.kirjasto.oulu.fi/tekstiversio/plain.html.

• Allow access to the full site. Some people may not want to visit only the mobile-optimized site from their device but would rather have access to the full one; it’s wrong to force a user to only access the mobile version. This is even more so if you break the other rules. It should, therefore, be possible to easily access the full, standard site via a link from the mobile site. Google does this well by putting a link in their mobile version that allows you to access the regular version. Many others have still not gotten there, even though it’s an exceedingly simple thing to do.

Examples

Let’s take a couple of examples of mobile-optimized news sites and see how they rank according to the above criteria. To start with, consider one of the most popular news sites, CNN.com. When you access http://www.CNN.com from a mobile browser, you’re directed to http://cnnmobile.com/. So far so good, considering the basic CNN main page is quite heavy. But afterwards there’s nothing but disappointments to be found here: there is very limited amount of content available, it’s spread over too many pages and way too much of the space is taken up by ads. CNN is a prime example of how NOT to create mobile site. Also, there’s no way to access the full version of the site from the redirected site – to do that, you have to manually enter http://edition.cnn.com/ – not very intuitive. Anyway, take a look at the crowded screenshots below – there is very little useful information on any one screen as the page is filled with other crap:

BBC News has a few versions optimized for different devices, maybe even too many. There’s the full-blown version at http://news.bbc.co.uk/ and a low-graphics version at http://news.bbc.co.uk/text_only.stm . Then there’s a PDA-version at http://www.bbc.co.uk/mobile/pda/, a mobile optimized version with images at http://news.bbc.co.uk/nolpda/ukfs_news/hi/default.stm and without images at http://news.bbc.co.uk/low/english/pda/default.htm – which one of those is one supposed to access?! It’s a trial and error process and not a very convenient one at that. If you access the main website, http://news.bbc.co.uk with an S60 mobile browser, you get to http://bbc.co.uk/mobile, which is a crappy WAP version of the news site with few and tiny images and no way to get to the real version. Not good. See below for screenshots:

A better example is NY Times; when accessing http://www.nytimes.com/ with a mobile device, you’re taken to http://mobile.nytimes.com/ which is a pretty usable mobile version. The advertising (when there – there are very few ads) is not intrusive and the pages contain plenty of text and stories – I’m not sure if it has all the textual content that is on the regular version, but it certainly has enough. The stories are split to pages, but the pages are reasonable in their length so that’s okay. Moreover, “paging” is just the default setting that you can change, another plus. There are also some images which is nice. Something I’m not, however, able to do is access the full version with the mobile browser.

Overall, many sites do some things right – like directing automatically to a mobile-optimized version without having to memorize a mobile site URL, but also some things wrong – like not (easily) allowing access to the full version of the site. Of the news sites, the CNN mobile version is quite useless while the NY Times one is already pretty good; BBC News falls somewhere between the two, but only if you access the right mobile version which you have to choose yourself – for S60 browsers, the one at http://news.bbc.co.uk/nolpda/ukfs_news/hi/default.stm works best IMO.

One more potentially lethal misstep that should be mentioned is content adaptation / mobile browsing gateways by the operators. They might work to some limited extent for very simple browsers, but some operators are planning on using these gateways to e.g. inject advertisements. A very bad idea. If you find a website that doesn’t “mobilize” well for your device, you can always use operator-independent mobile optimizers; Opera does that, Mowser used to do it and Google does it also. With increasing number of websites mobilizing reasonably well and as you can use operator-independent services for others, why should you tolerate an intrusive operator-run content adaptation gateway?

Btw, Google also does some decent site “mobilization” if you need such a feature – and finally, it might be worth to check out their other mobile services from http://mobile.google.com/ also.

While evaluating a course on Web 2.0 ecosystems and listening to stuff I already know, I was looking at other wonders of the web, namely news. Unsurprisingly after recent developments, there appears to be a not-so-slight market crash going on as we speak. Incidentally, it’s not the only thing going down: check out the exciting front page of NASDAQ:

The NYSE site is holding up a bit better – for now – but also revealing that at least in the very beginning of the day, the massive 0.75% rate cut by the Feds didn’t really do much to help soften the blow at the US end:

This ought to become an, hmm, interesting, week. Or month. Or a year.

Continuing on a somewhat related topic with “Freakonomics” that I read a short while back, The Undercover Economist by Tim Harford takes aim at partly similar things but from a different perspective. While the very beginning is a little bit annoying in its pomp, the book quickly gets under way to handle more interesting issues. A lot of interesting issues to be exact.

Like with most excellent books, it’s impossible to give an overview of what exactly is discussed – I wrote many pages of notes and there’s no way to cover even 10% of those in a short blog post. To give a quick glimpse on to some of the topics in the book, however, below some selected items I found insightful and/or fascinating.

• The hefty premiums paid for Fair Trade coffee and other goods is basically a scam, which is why many companies have begun to charge the same for Fair Trade goods than “regular” goods. This would be a lesson to learn in Finland, too. Anyway, over 90% of the extra price the consumer pays never reaches the farmer who is intended to be the beneficiary. Instead, it goes to the companies:
  

  .. so the money went to Costa’s bottom line – unless using fair trade coffee somehow increased their costs hugely. The truth is that fair trade coffee wholesalers could pay two, three or sometimes four times the market price for coffee in the developing world without adding anything noticeable to the production cost of a cappuccino, because coffee makes up such a small proportion of that cost.

  Sounds similar to my earlier musings about bread and rising wheat prices, no? The raw ingredients make up such a tiny proportion of the end product cost that justifying price hikes by increased raw material costs is nothing short of dishonest. The same is true for organic goods to some extent; for example in the UK, organic milk commands a 15p premium per pint, but the farmer sees less than 6p of this.

  Another interesting bit of information was to see just how little all kinds of “options” add to the production cost – compared to what they add to the final price..

• Ever wonder how people are lured to buy the more expensive stuff even when the cheap would be just as good? A lesson from supermarkets gives you some hints:
  

  Supermarkets will often produce an own-brand ‘value’ range, displaying crude designs that don’t vary whether the product is lemonade, bread or baked beans. It wouldn’t cost much to hire a good designer and print more attractive logos. But that would defeat the object: the packaging is carefully designed to put off customers who are willing to pay more.

• People often complain when health care services have to make value judgments on life and death matters because money is limited. Yet, people themselves constantly make value decisions; it’s enlightening to see just how often and how serious those value judgments can be even without us explicitly thinking about it. Consider this:
  

  Individually, we constantly make decisions that put a value on our own environment, our own time and even our own lives. If you pay more to avoid a noisy area when you rent a flat or a hotel room, then you have implicitly put a value on peace and quiet. If you decide you can’t be bothered buying a smoke alarm, you have traded off saved time and expense against an increased chance that you will die. However, when you make any of these decisions, you probably don’t come clean to anyone, even yourself, about the price you’ve put on quiet, time or life.

  This line of thinking leads us health care and why it’s so broken in so many countries. The US system, driven by problems with asymmetric information, is incredibly expensive and wasteful, yet doesn’t provide the results it could be expected to. A suggested solution involves keyhole economics with two guiding principles; ensure the widespread availability of information and give patients an opportunity to use this information. The proposed plan of catastrophe insurance and mandatory health care savings account sounds great – and what’s even better is that it’s been working in Singapore for decades. The cost? The cost of the system is $1,000 per person per year – less than the cost of bureaucracy alone in the US. Yet, the average age of Singaporeans is 80.

• Interesting insight is offered on two countries in particular; Cameroon and China. The former focuses on why exactly Cameroon is so poor. On China, the focus is on how it grew to be such an economic powerhouse. Both are fascinating accounts. On a lighter note, consider this on Cameroon:
  

  .. the most popular way to make the trip is by bus; minibuses ply all long-distance routes in Cameroon.Designed to seat ten people in comfort, they will depart as soon as thirteen passengers have boarded.

• Other fascinating topics included, but were not limited to, discussions on problems of game theory (mainly the assumption of hyperrational players), details on the 3G auctions and why some worked some didn’t, bidding strategies, the ecological benefits – or the lack thereof – of local production (local movement doesn’t translate to small transport costs – the transport costs of moving a CD player from Osaka harbor to the port of Los Angeles are less than the transport costs of moving it from the LA harbor to a local Best Buy branch), market failures vs government failures, taxation of imports and the free market. In other words, a very interesting collection of topics.

The Undercover Economist is constructed interestingly in terms of “plot” development; starting off from a few simple examples of things like coffee prices, it goes on to explain some key concepts like scarcity power and what Harford calls the “world of truth” that a hypothetical perfect open market would implement. Towards the end the book veers to handling bigger things like globalization and whether it’s good or bad, wrapping up with a chapter on the economic wonder of China.

As has been seen elsewhere – and as was seen in Freakonomics – what really matters in how people behave is their incentives. If you have poor, misplaced or misaligned incentives, you’re asking for trouble whether it’s on a small scale like in a coffee shop or large scale like running a country. While it’s a great book, the undercover economist is not a perfect one; for example, Harford gives some wrong information on why oil is so expensive. He also repeats music industry propaganda without any filters and states that we’re “probably not on the verge of another depression”, which is pretty close exactly what the US was a year or so ago.

Still, even with its shortcomings, I can highly recommend this book. It’s an enlightening – and at times even funny – book that sheds some much-needed light on many, many important issues. The list above doesn’t even begin to scratch the surface.

I just ran into a new experimental service that looks and feels quite nice: HIIT (Helsinki Institute of Information Technology) is a running a neat little project called Capricorn looking at adaptive user interfaces utilizing a standard browser. What’s more, it’s mobile-optimized which makes it a joy to use on mobile phones where it’s more suitable to begin with.

The idea is that you create a personalized portal of sorts by adding your favorite services; select from an existing list or add your own. This list is then presented as icons as can be seen on the right here. The gist is that the interface is adaptive; first, less-used icons are greyed out for easy deletion. Second, they have S60 software that gathers “environmental data to customize the interface and services” and customizes the services based on that information – presumably stuff like location (didn’t get a chance to try this out yet).

Clicking on the icons will take you, depending on the site, to a mobile-optimized version as can be seen on the right here, or to the original website. The top frame is populated by icons that will take you back to your Capricorn homepage.

I can think of a few improvements to the UI, like not having all the “logout”, “help” etc icons at the top taking up space – maybe just one slim “go home”-button. Moreover, the “Close”-sidebars that appear when you navigate to the edge of the page are unnecessarily intrusive and annoying. The +/- icons are also excessive, considering at least the Nokia browser already has single-key-press font size settings.

But anyway, such UI experimentation is very welcome thing indeed. I can easily see this developing into a serious competitor with Widsets.

Note: the service is in something like a beta test, but I’m sure they won’t mind a couple of extra users Register at http://universe.hiit.fi/capricorn/ to become a user.