Ada Lovelace day - Aleks Krotoski

As today is Ada Lovelace day I'm responding to the request on the Finding Ada website to "Write about a women in science, technology, engineering or maths whose achievements you admire".

I've chosen Aleks Krotoski. She's just good at so many things:

• She's an academic. She's got a PhD and when she writes about things it's pretty well researched and not just a blogger's opinion.
• She's good at broadcasting, including the BBC's Virtual Revolution programme, Radio 4's The Digital Human and the Guardian Tech weekly podcast. As those are group efforts I don't know how much she is involved in the compilation of these programmes, but if the stuff she's written with her...
• ...journalistic skills in the Untangling the Web column is anything to go by, a lot of it is her work. As the research is done in plain view on her blog it's interesting to see what's gone into making up the final words.

Anyone of these would be admirable, but to have all three in one person is more so.

Federated Identity and Identity Assurance - why you should care

Yesterday I read a blog entry from the UK Government Digital Service about Identity Assurance.

We’re helping develop a secure service that lets people log in to online government services more easily.

They link to this article by the Telegraph which describes it well (apart from the headline).

I've blogged before about the Polder Consortium where we're thinking about such things. Identity Assurance is part of Federated Identity. If you're in IT then I expect you're going to get friends and family asking you if it's OK to use your Facebook login to access government services. (I suspect a Facebook login may not reach the required standard of assurance.) I'm personally encouraged by the fact that they are consulting with people who worry about privacy.

I think, though, that if you're in IT it's worth understanding about federated identity, authentication, authorisation and assurance levels so at least you can have an informed opinion. (I'll point to a document that's soon going to become available when it does.)

In other news, as they say, the Polder Consortium has released some new standards, recommendations and notes, either because they've gone to Proposed state or because we've decided that some drafts are worth making public.

Is looking stuff up on Google doing work?

(I was going to tweet this, but my thoughts started to fill more than 140 characters.)

While working at home one of my children made the observation "you're just looking stuff up on Google, that's not work".

I was actually looking stuff up on stackoverflow.com, or more precisely, finding out what technique the Google analytics tracking code uses.

I said that I was doing work. It made me think, though, that "looking stuff up on Google" is a useful skill. Not just looking stuff up, because we all do that, but quickly finding solutions to problems that you face.

A while back someone (and here my Google searching skills have failed me) wrote on their blog what they put on a job application form under "preferred programming language":

1. BBC Basic
2. Any language + Google

In my job I dip in and out of various languages - Javascript, PHP, ASP.Net - and for using those using Google, stackoverflow and MSDN is essential. I can't remember the arguments for PHP's strcmp function, but I just search for "PHP strcmp" and I'm there. So yes, "any language + Google" covers it.

For software development it's not very controversial to say that web searching skills are necessary. The same is true for other IT related skills, as IT people have often been the first to share their knowledge using IT technology. Fortunately using the internet no longer needs IT skills so many other professions have shared knowledge using the internet.

Is there a job where "looking stuff up on Google" wouldn't be a useful skill?

Edit: This gives me an excuse for a bit of a rant. I don't like it when someone starts off their article on subject "X" with a statement like, "Google has Y million results when you search for X". OK, so you increased your wordcount, but that's research any one of us could do. Lazy.

Evolving layouts for web pages

Seven years ago (surely a century in internet time) I posted a blog post which I won't summarise, but repeat here as it's short:

It's a big deal when a magazine or a newspaper has a redesign. Similarly with websites. It's something to do with the fact that it's easy to churn out issue after issue/page after page using the same layouts, fonts etc.

I would have thought that in the dynamic web world you could spend more time with your templates, spending more effort making them work harder for you, so each new page looks like an evolution, or a minor redesign, whereas you haven't put any effort into that page, you just put it in at the beginning. Does that make any sense?

Today, I've just read an article in Contents magazine, Made to Measure, which expands that idea, and is so much better written than I could do. It doesn't say that all the work can be done at the beginning, but with thought I agree with them that some work needs to be done for each article.

So if static templates are too limiting, but per-piece art direction is too costly, how else can we make scalable, sustainable digital publications that are beautiful and accessible? To find a middle path, we can take a cue from the art of the tailor.

What is a website exactly?

I've come across a blog recently called The Pastry Box project - "30 People Shaping The Web. One Thought Every Day. All Year Round. Sugar For The Mind.".

This recent post talks about interviewing candidates, and a standard question for them:

What's your favourite website?

If I were asked that question it might take me a while to answer. My thinking went like this:

Well, I use Gmail a lot, but that's my email, not really a website. Google Reader too. I use Librarything to keep track of the books I've read, but I think of that as my library catalogue, not a website. Flickr's where I store my photos, Facebook is... well Facebook.

I can't think of my favourite website, because the websites I use most are places for doing things. You could ask the same question about books, and I'd be able to give you an answer (apart from the Bible and Lord of the Rings it's probably Wintersmith actually). Books are more all alike than websites.

Is there something else that could give rise to head-scratching if you asked about a favourite, because it is such a diverse thing?

"What's your favourite activity?" maybe.

Playbook - 10 days in

Thoughts after 10 days with my Playbook:

It is slickly executed, but suffers from annoyances. It goes slowly at certain points, such as when you go into app world and click on the search box. It takes a while before the keyboard pops up. When you suspend it you don't know if a swipe will wake it up. There may be some logic to it, but I don't know what it is.

The app store, aka App World doesn't have a lot of what you want. For example, I couldn't find an app that acts as a satnav. The pricing is high too - $5 for angry birds, when it's less than $1 on the Apple app store. The online app store doesn't let you filter by device type, which seems to me to be crazy. When you create an app you specify which one of about 30 types of app it is, but you can't filter by those too.

There's no Adobe Digital Editions, which I was hunting for because I wanted to read ebooks from my library. However I did find the app from the company that manages ebooks for our county - Overdrive, and so that made me very happy - apart from the bugs in the app itself. It runs in the android emulator so it's probably hard to track down the cause of these bugs - the software, or the emulator itself.

I've released my second app, GPS to Grid Ref, which isn't rubbish, unlike my previous one. It is simple though - you click the button and it finds your position and displays it in UK Grid Reference format.

I've got a Blackberry Playbook - plus QNX nostalgia

I've got a free Playbook. I got it by writing an app and submitting it to the Blackberry app world. I did almost the bare minimum needed as I didn't want to spend too much lunchbreaks on it. I downloaded the box2d javascript code - a physics engine which runs on Flash or javascript - and adapted the demo you can see on that page so that the circles moved as you tilted the device. It took me a few attempts to get my app approved. I didn't sign it the first time, there was a bug which didn't show up in the emulator (but did in the simulator - I now know there is a difference) and then I had to make it big enough fill the screen (fair enough). Even though I was doing the bare minimum needed to get an app published (BubblePlay2) it did actually do something, even if it wasn't much of a game.

I was amazed to find that 80 people downloaded it on the first day and 50 on the second. I guess it appeared on a "new apps" page. After that the downloads dropped right off. It got one 2 star review from someone who said it didn't work. The Blackberry simulator enabled me to check that tilting the device would move the circles, but I didn't know just what angle it was working at.

The Playbook runs the QNX neutrino operating system (a real-time OS designed mainly for embedded applications) which I came across in a previous job, so time for a bit of nostalgia.

QNX nostalgia

In the job before my current one I was working for an electronics firm. We developed one of the early webcams. It was a clever bit of kit - a PC104-based PC with a camera board that we developed. It ran the QNX Neutrino operating system. This was back in 1997. The company QNX already had a mature embedded operating system with the same name. Neutrino was their new product which had multi-threading and we were a beta customer. I wrote a point enabler for the card, a cut-down device driver designed to work with one particular device - the ethernet PCMCIA card we were using for networking. I flew to Canada for a week to the QNX offices near Ottawa to pick their brains.

Canada

I arrived at the airport on the Saturday evening, so my first experience of driving on the right was to drive a hire car through the dark into an unfamiliar city. Fortunately it was built on the typical North American grid pattern and once I'd worked out what the one way signs looked like I managed to get to my hotel.


It was February and the temperatures were well below freezing. People were skating on the Rideau Canal. I fancied having a go, but I didn't want to fall over and injure myself when I was supposed to be working. I drove around the city on the Sunday and also experienced walking around in -20C temperatures.

A colleague of mine flew over early in the week. He was helping QNX debug their TCP/IP stack as we needed to get it working and we kept on finding bugs. We both got our jobs done in a week and I got to see Canada.

Our webcam wasn't successful. At around £2K it was quite expensive and although we had ideas as to what you could do with a PC onboard a webcam, we didn't find a market for it. After I left the company I heard that we had loads of them stockpiled in one of the offices. Which brings me back to the Playbook. My theory is that one reason that RIM are giving away Playbooks, is because as they've written lots of them off on their books, they have an asset value of $0 and so it's worth giving them away to try and increase the number of apps on their platform. Even if some of those apps aren't very good, like mine.

Photo from Douglas Sprott (cc) some rights reserved.