Alex R. Young

What Was The Future of Web Apps?

2012-12-05T00:12:00Z

I went to either the first or second Carsonified Future of Web Apps/Design conferences -- I think it was back in 2005, so over seven years ago now. I've been thinking a lot about that time and the things that used to preoccupy me: 37signals, Ruby on Rails, Digg, Apple products, Ajax, and IE bugs spring to mind.

What happened since then?

The most obvious answer has to be "the cloud". Historically, distributed systems and software/platform as a service have existed since before 2005, but Amazon launched AWS in 2006 and that's when the term started to become widely known. At least that's when journalists started using it in an annoying way.

Ajax exploded, JSONP became the norm, client-side JavaScript evolved from something evil to exciting and finally accepted. jQuery rules the roost. There are those working to break client-side scripting into smaller, more manageable libraries -- the Component and Ender spring to mind.

XML RPC is now largely regarded as a painful corporate technology, while RESTful JSON APIs are far more common on the contemporary Web.

Authentication has been democratised by OAuth -- I know I'm more likely to try something out if I see "sign in with Twitter" or GitHub.

Speaking of Twitter... It's amazing to think about that first conference existing without Twitter. Twitter appeared in 2006, and has been prominent at many related conferences since then.

2007: iPhone. At first Apple claimed applications would be purely delivered through Safari, but eventually the App Store appeared and changed everything. Although industry stalwarts like Ericsson, HP, Palm, Nokia, and even HTC (they had early touchscreen phones) were all making smartphones, Apple's move away from physical buttons stole everyone's imagination. I'm not sure I'd have even used Twitter at conferences without a smartphone.

Hacker News: 2007. I use Hacker News (and reddit) purely as a way to find interesting content to read. One thing that I like about Hacker News is it often feels like programmers trying to hack business -- a pool of shared experienced to help the wider community conquer business-related issues. That community has done a lot to sell entrepreneurialism to fellow nerds.

Heroku: 2007. Initially aimed at Ruby developers, Heroku simplified web application deployment. Thinking back to 2005, it's something I definitely wanted at the time, and I had my own ridiculous cocktail of scripts that did something similar. I'm still running web applications on servers that originated in that era, and as a lone developer (with a network of helpful friends) I wish I was running my apps on Heroku or similar services.

In 2008 GitHub was launched, and is probably one of the only companies I've consistently wished I worked at. Before GitHub, most of us struggled along with Subversion and sites like SourceForge, but it's fair to say GitHub has revolutionised collaboration for programmers and web developers in general.

For me 2009 is when everything started to change. My interest in JavaScript grew to the point when I felt the need to start writing DailyJS, which I still enjoy running today. In the same year Node was released, and coincidentally Redis. Both had a focus on performance, addressing issues that were hinted at years before but became more widely known around this time. The main issue being I/O as the major bottleneck for web application performance. MongoDB was also released in 2009, so this was really the nexus of NoSQL.

I remember Blaine Cook's talk about using XMPP to make scalable web applications back in 2008. The concept of publish/subscribe has been embodied in many popular technologies since then, and event-based programming in general is pretty fundamental at this point.

I'm aware events is not the only solution. I enjoy my share of Go and Clojure, but the production realtime software I've made has been powered by Node.

After 2009 apps started to take over. I think if I ran a conference like The Future of Web Apps I'd have wondered if native smartphone apps were a threat to the potential audience size. I still seem to meet far more developers who write Ruby, Python, or use Microsoft's web development technologies than write Objective-C or use Android's Java SDKs. I went to an iOS developer conference in 2010, didn't meet many iOS developers but landed a contract from someone looking for Objective-C programmers.

As a web developer who also writes "native" code I saw apps as an opportunity, and it forms a healthy part of my business as a freelance programmer. In 2012 it felt like an equal share of startups on Hacker News were working on web applications or mobile apps (or both).

Around 2010 is when responsive design started to become prominent. Designers were having to support multiple types of mobile devices, including the iPhone 4's Retina Display. Resolution independence is something that's been around since TeX, but the reliance on bitmap graphics in desktop and web design makes it difficult to use everywhere. Responsive design has given us web sites that are able to serve different images (or totally different layouts) based on the device's resolution. This is a good thing, but I wish designers would figure out the difference between the brand name "Retina Display" and the concept of supporting high DPI displays in general.

Since 2005 a lot of things have changed for the better: HTML5 seems to have caught on, proprietary technologies like Flash and Silverlight have received a steady drumming, Git lets me work offline and collaborate without many headaches, Node lets me write JavaScript on the server in a way that I enjoy.

I don't think the Future of... conferences were ever meant to be prescient, but I feel like web apps still haven't met their full potential. Yes, small ARM computers and 3D printing are exciting, and I anticipate exciting events to occur in the hardware hacking space. However, I'll still be juggling hundreds of usernames and passwords, web applications that can't talk to each other intelligently, and vendor-specific rendering bugs.

Here's to the future of web hacks!

Silent Running

2012-11-07T00:11:00Z

I have a limited edition Masters of Cinema SteelBook edition of Silent Running. I'm not a big collector of films, but Silent Running is special. This particular edition has been beautifully transferred and includes the The Making of Silent Running documentary.

Silent Running was introduced to me as a child by my father. I was young enough not to notice the age of the film, I had no idea that it was made before I was born. I liked the robots, space adventure, and strange mix of nature and futurism. As a teenager I reveled in the sounds of the movie; not just the score but the incidental sound effects. There's dialog from Silent Running on DJ Shadow's Entroducing, which I picked up on the first time I heard it. It wasn't easy for me to see films at this age, because I couldn't afford a VHS copy so I had to hunt around on the limited terrestrial TV channels. I had a friend at school who'd excitedly inform me when he'd spied Silent Running in a TV schedule.

As an adult it's impossible to ignore the film's ecological message. The Earth is dying, so enormous spaceships have been created that hold pockets of Earth's ecology. Although Silent Running was released in 1972, the concepts behind it were developed much earlier.

Growing up in the 80s, global warming was taught to us in class even at primary school. I think it was only mentioned in passing, but the link between carbon dioxide and climate change had been understood for decades already. I dimly remember TV news and documentaries showing that NASA scientists had used satellite data to prove global warming was happening. This bothered me a lot, and I remember creating designs for solar-powered street lamps and sending them to our local electricity board. I used to be embarrassed about that, but given my chosen career it seems pretty normal in retrospect.

As a child of the 80s, I assumed that environmentalism and climate change in popular culture was cutting edge Space Age knowledge. I knew that 19th century geologists were aware that the climate had changed, but environmentalism goes back hundreds of years. In 676 Cuthbert of Lindisfarne enacted laws to protect seabirds nesting on the Farne Islands. The WWF was registered in 1961, yet John Muir had already founded the Sierra Club in 1892.

I won't pretend that growing up alongside convincing proof of global warming in popular culture hasn't shaped my opinions about climate change. However, wondering where a science fiction film's environmental message took me on a fascinating journey through the history of science. Silent Running wasn't ahead of its time, it's just global culture is still behind the times in its continued destruction of our environment.

The Open Tsunami

2012-11-04T00:11:00Z

I had a G3 iBook (PowerPC 750FX, mid 2002), and buying it stretched my finances to say the least. During my university years I became an open source zealot, using FreeBSD and Linux on my computers and enjoying it immensely. I bought a Mac because I also liked writing music, and there was a much richer selection of commercial software available for Macs. Upon buying the second generation iPod, I considered buying shares in Apple because it seemed like they had a perfect storm on their hands: Unix, incredible design, and products that appealed to both the non-technical and hackers among us.

I really should have bought those shares!

That iBook came with me to several technical conferences around London, and Macs weren't yet particularly popular with my fellow programmers and designers. They were probably more popular in more affluent technical communities around the world, but programmer salaries have never been particularly great in London. Back then owning a Mac garnered something like a motorcyclist's wave: the Mac-owner's nod.

The iPod captured my imagination. Slashdot hated it, but I loved it because of the interface. I thought the future would be an iPod-like machine that replaced laptops and computers: just dock it with a portable keyboard and screen, or a larger screen and keyboard at a desk. That never really happened -- the Motorola ATRIX wasn't a bad stab at it but I always felt like booting a secondary OS for the laptop mode seemed weird.

It's pretty clear that Apple have dominated developer mindshare. By switching to Unix, they won the hearts of developers who were frustrated with Microsoft or with maintaining Linux. Apple's products promised to save time, and delivered unparalleled hardware design. Eventually, those same conferences were packed with Macs. I recently went to a Vim meetup in London, and Apple laptops even dominated that audience.

Prominent tech writers have publicly switched from Macs to an open source OS: Mark Pilgrim and Cory Doctorow seemed to be at the head of that movement, back in the days of Digg and the Battlestar Galactica reboot. People looked on with interest, yet continued buying Apple products.

Since the launch of the App Store for Macs, and Windows 8, it seems like the future of computing is a locked-down store front optimised for advertising and sales of software and services. All of the major players have services for cloud storage, music, movies, and TV shows.

Developers infinitely more intelligent than me see this as a move towards a draconian, closed future:

"They get to certify programs and control the experience. This is great for them (and possibly arguable makes for a smoother end user experience as well, but that's debatable if it's good), but it places faaaaar too much power in the hands of a single entity."

Notch, from John Callaham: Don't be a goat murderer

On one hand, Apple and Microsoft are making it easier for me to make money. On the other, their certification represents an insidious attempt at controlling what software you're allowed to run on your machine. Of course, the concept of ownership in the hardware, software, and media industries is more akin to licensing. DVDs are licensed and cannot legally be used in certain ways -- it seems like we have equally restricted rights when it comes to physical discs or digital purchases.

The disparity between licensing and the sale of a product is underlined by transfer of ownership. When using an app sold by Apple or Microsoft, you're entering into several licensing agreements where the manufacturer still technically retains ownership. We're living in a world of rented ideas.

Therefore, whether software is purchased shrink-wrapped on a disc from the high street, or through a service like Steam, the effect is legally the same. However, one important difference lies in termination. The Steam Subscriber Agreement states the following:

"Valve hereby grants, and you accept, a limited, terminable, non-exclusive license and right to use the Software for your personal use in accordance with this Agreement ..."

"Valve may terminate your Account or a particular Subscription for any conduct or activity that Valve believes is illegal, constitutes a Cheat, or which otherwise negatively affects the enjoyment of Steam by other Subscribers."

Valve has the rights and means to prevent you from playing your library of purchased games. If you unwittingly broke the terms, Valve could prevent you from accessing every game you've purchased through the service.

We've arrived at a curious point in time where we've allowed the following things to happen:

Apple laptops are ubiquitous at developer events,
Apple, Microsoft, and third party services have the ability to censor software and restrict it,
The media and software we buy through these services can be taken away from us.

The Imperfect Storm

Much like Apple's perfect storm of hardware and software ten years ago, we're seeing a new movement that will rise against Apple and Microsoft's newfound success. This movement is incredibly cheap, hackable, and open hardware, that will generate a resurgent interest in Linux and other open source operating systems and software.

Processing was launched in 2001, which made software art projects more accessible to artists. Arduino was launched in 2005 as an open hardware and software project to make microcontrollers more accessible. Arduino makes programming microcontrollers easier than writing C, and more accessible than typical electronics projects. The creators shrewdly reused Processing as the Arduino IDE.

Perhaps inspired by the Arduino maker movement, early concepts of the Raspberry Pi were built on the Atmel ATmega644 microcontroller. One of my early programmer heroes was David Braben, who was one of the founders of the Raspberry Pi Foundation. This charity was founded in 2009 to promote the study of computer science in schools, but the Raspberry Pi has been hugely popular with makers, hackers, and tinkerers.

A more casual developer who loves Apple's hardware may scoff at this, but it would be a mistake to underestimate how popular the Pi and Arduino will be within the next few years. As programmers we're hungry for things to make: the need to create drives us to learn programming languages in the first place. I see more questions about Linux on the web than ever before, and this is partly down to the interest in the Pi.

This new grassroots movement is going to be one of the things that topples Apple. Back in the late 90s, many of us became interested in Linux due to frustration with Windows 9x. Over a decade later, renewed interest in Linux and cheap, hackable hardware will eventually branch out into more open hardware. Developers need laptops and desktop computers -- consumers may be happy with tablets, but we are not.

The Raspberry Pi Foundation has shown that developer-centric hardware is economically viable, GitHub has shown that developer-centric services can be successful, and Apple have shown that developers appreciate well-designed hardware. System76 sell computers for Ubuntu, but expect more of this in the future: the trickle of developer-friendly gadgets is going to turn into a tsunami within the next decade.

Eighteen Months

2012-10-24T01:10:00Z

"I like your computer," she said. "It looks like it was made by Indians or something."

Chia looked down at her sandbenders. Turned off the red switch. "Coral," she said. "These are turquoise. The ones that look like ivory are the inside of a kind of nut. Renewable."

"The rest is silver?"

"Aluminum," Chia said. "They melt old cans they dig up on the beach cast it in sand molds. These panels are micarta. That's linen with this resin in it."

From Idoru, by William Gibson

I've been trapped in the 18 month hardware cycle for over a decade. Whether it's mobile phones on a contract with a "free" upgrade, PC hardware replacement driven by games, or Apple's relentless yet undeniably compelling product refreshes, I can't escape it. I won't lie and pretend that I don't get excited by Apple keynotes or video game hardware news, but we can blame Moore's Law for a certain amount of this thirst for improved technological artefacts.

"David House, an Intel colleague, had factored in the increasing performance of transistors to conclude that integrated circuits would double in performance every 18 months."

Wikipedia, Moore's law

Every time a major innovation breaks into widespread consciousness, people ask if we've reached the end of Moore's law. However, several trends outside of raw hardware performance now threaten Moore's law more than anything else. Rare earth metal exports are being restricted, so previously closed and new mines are being opened to supply the huge demands of the electronics industry. There have been calls to take better care of this resource, and I expect electronics recycling will become a key part of recycling programmes alongside the now ubiquitous glass and paper collections. I noticed in suburban Japan trucks drive around at least once a week collecting old electronics and household appliances, in the UK we have to take such goods to specific refuse collection locations.

Another trend is crowdsourcing and maker culture. There's a huge amount of interest in the Raspberry Pi, Arduino, 3D printers, and the weird and wonderful products appearing on Kickstarter and Etsy. More people are becoming interested in solving their own problems -- I expect a homemade laptop/tablet movement to kick off if it hasn't already. Put a Kindle and a Raspberry Pi together and you've got a portable machine with incredible battery life. Or, just root your Kindle to use it as a general-purpose computer. I'd actually love an eink laptop with a mechanical keyboard: a portable machine purely designed for writing. It's unlikely that anyone will make this commercially, but I don't need them to because I can make it.

Mobile CPU and graphics hardware will continue to follow Moore's law for the foreseeable future, but many people are satisfied with existing hardware. At one point mobile phones were too big, LCD screens were poor quality, but those are solved problems. I'm extremely happy with my launch iPhone 4, and although I have a lot of devices purely for testing the apps I build, I don't see any reason to upgrade to the iPhone 5. My phone is in excellent working order, although at over two years old the battery will start degrading soon. Apple are notorious for including fixed batteries, so there's going to be a point where it's easier to get a new device than replace the battery. This built-in obsolescence has spread to Android as well, where certain prominent flagship devices no longer support removable batteries or memory cards.

A big issue with pushing hardware beyond its intended lifetime is security. The first iPad didn't get iOS 6, Google quickly got bored of my Nexus One, but I could put community-supported software on these devices.

Etsy, or a similar site, may become the home of William Gibson's sandbenders. Send in your obsolete but well-loved device, get the tarnished casing updated, batteries replaced, and maybe even a community-supported software update. Although I'm still locked into the 18 month upgrade cycle, I like to donate old hardware to charity or friends and family.

If I was in a position to create an electronics company, I'd make the unique selling point "hardware for life" -- open source, removable batteries, removable storage, a mail-in service for replacing the casing and upgrades. The "stores" would be more like a hackerspace than anything else. This goes against the current model for an electronics company, but it might work as a non-profit service-based organisation rather than deriving most of the revenue from new hardware sales.

More Programming Languages

2012-10-17T01:10:00Z

We should not be supporting more programming languages. We should be killing them. All these bullshit projects are confusing people.
— ryah (@ryah) October 17, 2012

Picture Mark Zuckerberg sitting down to write an early version of Facebook. He knows enough PHP to make it happen, he's ready to hack all night to a Trent Reznor soundtrack, with enough Shasta to quench an elephant's thirst. At the end of the day he's tired, but chills out the best way he knows how: eating a burrito, drinking bad instant coffee, and browsing the latest news for nerds. He starts reading a discussion about how PHP is a terrible language, and questions whether PHP was the right choice for the nascent project that he's embarked upon.

He decides to rewrite Facebook in the current language favoured by the alpha geeks on the web. Do you think he'd still have built Facebook, or would the project have been derailed by an endless quest for technical perfection?

I don't care about the programming language used to build Facebook. The main point is Zuckerberg had an idea and implemented it using tools he was comfortable with. Similar stories can be found in the game development community: modders start off by hacking a game they love, then move to Unreal Engine development, and eventually they're working on big budget games.

Gearbox Software started out by developing extensions to Half-Life, and have since created Borderlands and Borderlands 2 -- there are many more examples of developers that have taken this path. Although modding might not require the technical acumen of creating a game engine from scratch, Gearbox had ideas and found the shortest path to implementing them.

Thousands of XBLA developers are making games with C# -- it's amazing what people can do with an original idea for a game mechanic and enough programmer art to get the ball rolling.

Gamasutra has a good post about the art of Braid that shows Jonathan Blow's programmer art before David Hellman stepped in. The final game is beautiful, but the game worked because the mechanics were compelling. Jonathan could have given up development in a quest for the perfect programming language and the perfect art, but instead he got something interesting working then sought help with the weaker areas.

@infynyxx scala, clojure, coffescript, lua, C#, ruby, perl need to beeuthanized
— ryah (@ryah) October 17, 2012

"You're doing it wrong," said the joker to the thief. "There's too much confusion, I can't get no relief." Imagine if someone told Bob Dylan he was doing it wrong -- the world wouldn't be the same. That taciturn phrase is overused and betrays a lack of imagination: don't shoot down the enthusiastic youth with original ideas, but rather embrace and encourage.

Your programming language isn't your identity. It's pretty easy to learn more than one. You don't need to be an expert in all of them; it's fine to gravitate to areas you find aesthetically pleasing or more productive. I tinker with Lisp, I really enjoy Clojure. I make no money directly from these languages, and I'm not planning on creating a new company based around Clojure, but I find learning it enriching.

The multitude of languages enriches our wider culture as hobbyists, hackers, computer scientists, and engineers.

Can you imagine if an anthropologist decided that certain facets of human culture should be culled? The fallout would be spectacular. The obvious argument against what I'm implying here is that programming languages are tools, and don't warrant the same level of cultural importance as art, literature, and science. However, this is purely dependent on your point of view. I think anyone with a passing interest in programming or related technical fields could identify the cultural underpinnings of the major programming languages.

Prominent bloggers, including Zed Shaw on Shedding Bikes, discuss programming culture. Zed moved away from Ruby development partly due to cultural reasons. So culture is important to programming -- the flavour and style of a language is only partly derived from the language creator's design. Languages are often created to solve a particular need, but as the language becomes more generalised and capable the library authors and technical writers start to shape the culture around it.

This isn't an attack on the author of the above tweets. It's a plea to people who are about to embark on a creative project. If you have an idea, see how far you can take it with the tools you have to hand. If you've got spare time, learn some new tools. Don't feel like you're doing it wrong, you might be seeing things in a new way that proves to be fruitful.

The Technological John Peel

2012-10-10T01:10:00Z

When I was a teenager I regularly listened to John Peel's Radio 1 show. Peel is credited as breaking entire genres, as well as bands. He was one of the first broadcasters to play progressive rock in the UK, and he championed bands including The White Stripes and Smashing Pumpkins who both went on to be hugely successful.

"Oh well, it's tough being 14." I know it is Clara, this is for you.

-- John Reads Clara's Letter, from the Peel Session for A Huge Ever Growing Pulsating Brain That Rules From The Centre Of The Ultraworld by The Orb

When rave started to reach public consciousness, Peel was one of the few mainstream DJs playing it. My father said something about listening to that "strange music where songs seem to last 10 minutes". This intrigued me -- I'd been listening to grunge for a year or two so all I knew was guitar bands and pop music. We flipped the radio on after a summer barbecue and Peel was playing The Orb. It left a lasting impression on me, and to this day I still listen to the same tracks.

As a teenager I naturally gravitated towards certain musical genres because I was trying to find my own identity. It's socially efficient to identify with a specific group. I was originally interested in rock music, but went on to discover rave, ambient, drum and bass, progressive rock, house, techno, and more. Before long my CD collection had grown into a decidedly schizophrenic mess.

Whenever a hardcore fan of a particular genre would criticise or question my interest in an opposing musical enclave, I'd think about John Peel. He never seemed to care about what kind of people listened to a certain type of music, what clothes they wore, or what label the band was signed to. He just had a genuine interest in music.

Fast-forward to the textual battles of contemporary technology culture. Android or iOS? Macs or PCs? Vim or Emacs? BSD or Linux? PS3 or Xbox? Python or Ruby? Node or Go?

Technology culture has its roots in entrepreneurialism, science, and occasionally the arts. For our industry and culture to evolve, we need a generation of technological John Peels. When I think about the contribution John Peel made to contemporary music culture in the UK, I can't begin to imagine what the equivalent in my field could do.

DailyJS: Sponsored Content

2012-09-12T01:09:00Z

"Money is only a tool. It will take you wherever you wish, but it will not replace you as the driver." -- Ayn Rand

I'm trying out an experiment on DailyJS called Sponsored Content where people offering commercial products can pay to get an article written and featured alongside the usual content we publish. That almost sounds like payola, so it's going to take care and tact to manage it.

My goals are fairly straightforward:

Clearly distinguish between commercial products and services, and the usual (generally) open source things we cover
Use the money to pay people to write for the site
Generate money to free up my time so I can create more sites like DailyJS

Let's say you make an API or service that's relevant to JavaScript developers -- perhaps an error notification service, or supercharged CDN. By paying me to write about your service, you'll get a tutorial and prominent placement on the site. I see it more as a combination of genuinely useful content and advertising than simply yet more ads blasted into eyeballs.

The author of the first commercial product that we featured was extremely happy with the results: RoyalSlider: Tutorial and Code Review. It's partly a tutorial but also a code review, in which I pull the code apart to verify the quality. It turned out to hold muster, so I was more than happy to feature it on the site under the Sponsored Content Programme.

Put Your Smallest Clients First

2012-07-13T01:07:00Z

I've been freelancing since 2005, which means I've been doing it just a shade longer than regular employment (whatever that means). At this point I feel somewhat qualified to talk about how to survive as a freelancer, although the process has been made less turbulent for me thanks to a handful of regular clients that keep my baseline income stable.

That leads us to point number one: look for stable contracts. Corporates are looking for contractors all the time to support technical projects that they don't think are long-lived enough to warrant permanent hires. However, from the freelancer's perspective, their short timescales are quite long (six months, which can easily lead to several years). And the best thing is a reasonable hourly rate is cheap compared to a corporate salary (plus tax and benefits), which means from our perspective as lowly freelancers we actually get a good deal.

Still, being on a contract doesn't guarantee regular work. If you're on a retainer your contract might forbid other clients, but in the UK I'd be wary of this because ideally you want to run a company rather than be considered an employee according to IR35 rules. It's particularly important for me to avoid being seen as under "disguised employment" because I invest some of the money I earn back into my own entrepreneurial projects.

Naturally this means taking on multiple clients, not only to supplement income but also to avoid the IR35 income-draining spectre. More clients leads to less time, so management becomes more crucial. That brings me to my second point: put your smallest clients first. Smaller clients need as much attention as larger clients in terms of social engagement and support, but obviously the actual production of their software is less intensive. The pull of larger projects can be distracting, making it easy to forget about the smaller ones.

Avoid this by starting each day by working for your smallest clients first. And keep them in the loop -- email regularly, make videos or even just screenshots to demonstrate the latest features. This type of asynchronous communication is faster than a conference call.

Another thing to remember is this: you're a freelancer. That means you're free to take the afternoon off, or go and read in a café for an hour. Don't limit your experiences by structuring your entire day around work, else you might as well get a regular job.

To summarise:

Look for stable medium or long term contracts to maintain a sensible income
Put your smallest clients first, don't let your quality of service slip -- but limit the time you spend on their projects
Remember the "free" in freelance

Researching for Technical Writing

2012-05-25T01:05:00Z

How are technical articles and books written? It all comes down to research. Every editor I've worked for has always extolled the importance of research, and a well-researched article is certainly more pleasant to write.

The following suggestions are based on the methods I employ intuitively, so I may have missed out some important points. However, I hope it gives readers some ideas on how they can bolster their own technical writing.

Reader Competency

First, decide who your article is for, and what they already know. Writing for beginners means drawing on standard documentation, and perhaps putting more creative effort into example code.

Writing for more advanced developers requires more work. When targeting readers at this level a huge amount of effort should go into understanding deeper issues. For example, rather than researching core libraries, I'll research more involved problems such as structuring or scaling large applications.

Community

The community around a given technology generates a wealth of research material. Try reading mailing lists, blogs, social networks, and Hacker News/Reddit comments. This really helps show what areas commonly confuse people, and therefore what might be worth writing about.

When researching Windows and Node: Portability, I found several relevant posts to the nodejs Google Group by prominent developers in that community.

Issues

GitHub and similar services provide issue tracking. Bug reports and feature requests give amazing insights into the problems users have.

It can even be productive to see how the authors fixed bugs, because it might demonstrate an interesting technique or a weakness of the underlying technologies.

Related Projects

Don't just look at the core technology that you're covering, but also read about prominent related technologies.

If I wanted to write a book about Ruby, I wouldn't just research Ruby and its core libraries, I'd also focus on its major open source projects. Researching the Rails community would give big clues as to how people perceive and use Ruby itself.

Source

Finally, any time spent looking at the technology in question's source code is time well spent.

When I write about Node, I always have an editor open with Node's source code in it. This helps me expand on material in the standard documentation, and also learn techniques straight from the developers who made the technology I'm writing about.

Saint Spectrum Day

2012-04-23T01:04:00Z

Image from: wallpapers for World of Spectrum

I've been working as a programmer for 11 years now, and it's partly due to the ZX Spectrum 48K that my dad got me at some point in the 80s. I wanted to learn to program it, and the fact it booted straight into BASIC encouraged a lot of experimentation.

Computers don't work like that anymore, but some people want them to.

People find it surprising that there was an influential British computer, but the truth is there were many computers that had limited releases in their native territories; some of them used fascinating Russian Zilog Z80 clones and were compatible with Sinclair's machine.

Looking back, it's easy to think that the 80s saw a more diverse era in terms of home computer hardware. However, if you consider that the Z80 was actually an improved implementation of Intel's 8080, and was used in everything from the Game Boy to the Amstrad CPC, then it's debatable whether things have really changed in terms of hardware diversity.

The biggest change in computing since the era of the Spectrum is societal -- we decided to start connecting our computers together on a global scale, and putting connected devices in our pockets. Now computers really are a canvas for unrestricted creativity.