Normally, after a few years of *something* existing it’s time to party. Perhaps a re-design, re-purpose or simple “thank you” to all those who have followed the ups and downs and still continue to tag along for the ride.

If there is one universal constant, it is that I am not normal. Another is that what goes up, must (eventually) come down.

And it is with a combination of sadness (and hope) that I officially close down smackfoo.com. Whilst the site will remain active for posterities sake¹ it will cease to be my primary voice and will as such not see updates.

It’s also time to bid Wordpress goodbye. It’s been a fine vehicle for basic content management, but is simply not heading in a direction that I chose to follow. It is increasingly difficult to maintain code for the platform and consumes more time than it really should.

I have always appreciated the chance to share ideas and philosophies² and I thank each and every visitor and regular alike for dropping by.

As I’ve hinted at previously there is however, a new love in my life. The Lab has been quietly taking shape and has developed over a few months of mixing chemicals and powders together to see what explodes. The odd scorching aside it’s worked out pretty well.

Thus, I would like to invite you all to visit the Atomic Ninja Labs — my new home.

1. Wordpress themes and plugins will remain available. (↩)
2. not to mention all the discussion generated behind the scenes (↩)

I recently received an email from Joyent’s Kristie Wells, asking a pointed, yet perhaps fair question.

I wanted to reach out and see if I could ascertain the sour taste in your mouth for Joyent. I know you host with MT and S3. That’s cool, I have just seen more than one posting from you about not liking us and would love to know more. Feel like sharing?

Whenever I read an article or comment, that paints an open source product as though it was instantaneously better than closed source, simply because it’s open, it invariably results in one, simple question. Why?

The back-story here is that I have in the past questioned some of Joyent’s motives, where talk and action didn’t appear to align. It isn’t really a case of having a go at Joyent itself and it’s certainly not a “sour” viewpoint either, although I can see how it might appear that way.

Rather it is in reaction to the occasional belief expressed that suggests Open Solaris, Open Source and indeed their own company can do no wrong because they champion the open source way. If you promise the moon, you better figure out a way to do it before you start.

I should probably warn folks at this point. There are a number of ideals and considerations at play here, so this response is a lot longer¹ than I would have liked.

Mantra, cha-cha-cha.

To be fair, it’s not just Joyent that sometimes speaks in open-source is alpha-and-omega tongues. They have genuinely achieved amazing results with a combination of Open Standards and Open Solaris — they have every reason to feel proud of their achievements. The Accelerators are also a fantastic example of the potential of Open Solaris mated with a good application platform.

Indeed Open products and services are no longer the domain of the geek, hacker or 30-year-old-virgin living in their mother’s basement. We are all increasingly exposed to a raft of technologies built on open source or open standards code that, well, anyone can have a hand in building — again Joyent have played their part in that, too.

But old rivalries between the apparently ‘evil’ closed barons and the open source ‘rebellion’ die hard. Voices of the rebellion sometimes take aim squarely at closed platforms that actually provide reasonably open, documented APIs. Sometimes that logic can fall a little flat, particularly when environments fail.

Conversely, closed platforms aren’t entirely free of sin either. Vista proves that, certainly at an OS level, it’s not always plain sailing.

I’ll take that, to go.

It’s understandable then, that a product isn’t rated solely on how open it is, or more portable, rather it’s how well the very same product stacks up under real world conditions and whether the results are the same as the description on the label. It can even come down to the simplest equation of how easy it is to use. If you claim something special — you’d better be prepared for scrutiny if it doesn’t deliver.

Of course, there are two sides to any coin. The flip side is that the average user often just doesn’t care. The question isn’t often “what can your platform do for me?” — even if it is a valid question — it’s “will my stuff run on your platform?”.

So, if an Open platform can do that which the user asks, then great² but the risk is that if Closed-Product-A works when Open-Product-B falls over, then Closed-Product-A is considered better by default, even if an Open product is a better fit. The internal mechanics or proprietary risk aren’t always relevant as a result.

Benchmarks.

We’re repeatedly told that Open Solaris is better than the competition, “you don’t want proprietary lock-in” and they aren’t as open as us. It can and paint Joyent badly, even if unintended. Because it is in the same vein as “Trust us, we know what we are doing”. It’s dictatorial and makes assumptions, rather than actually giving solid reason.

That assumption can lead to a situation where failures just aren’t learned, communication falters, users are blamed and an eventual disconnect between the developer, supplier and the end user occurs. People don’t remember successes half as much as failures, they are the great leveller and prior claims will often be the meter used to judge an outcome.

It will invite (occasionally heated) debate, quite obviously.

Open vs Better.

I wouldn’t comment at all, were they not an application and hosting services company I have interest in and respect for. Whilst I host with (mt) and use Amazon’s S3 for storage³, the Accelerator product has always struck me as a solid framework to really build a good platform on. The problem however, is that the learning curve is steep.

A complex (yet open) environment isn’t always as tempting or useful as a simple point and click (yet closed) alternative. In the same way, I want something simple that will still scale, without that learning curve. There is actually huge potential in Joyent bringing a point-and-click type model to the market.

At some point the stability, capabilities and ease-of-use of a service become vastly more important in comparison to whether or not said platform is Open⁴.

Communication.

It really is okay to believe strongly in your product and in your business — Joyent have a great deal to offer start-ups as has been mentioned — but it’s also okay to listen to prospective customers and users, even if they don’t agree with you on some points.

I question to understand. Will my stuff work on your gear? Will you be there when I falter? Will you communicate when you falter? Will you learn from past mistakes? Will you stand up to scrutiny and take the time to listen? Can I measure you on the claims you make?

They are all vital questions, with answers that speak with far more authority than any “trust us” line can ever achieve. And Joyent should expect to be measured by the very things they claim to achieve — indeed they hold their competitors accountable in the same manner. Ultimately then, it’s not a case of simply saying you are better and having that stick. Rather it is the actioning of those words that counts.

The preparedness to listen, the encouragement to respond and to have those questions taken on board is a promising first step in putting action ahead of word.

1. Believe me, I have been editing this on and off for hours. (↩)
2. Transparency of code and open ideals can result in many positives, but that is frequently at a development, rather than user level. (↩)
3. Media Temple serve me well, keep me informed and in particular, don’t make assumptions and listen to feedback. S3 just works, it’s not exactly portable, but it is reliable and simple to deploy. (↩)
4. By choosing to use Open Solaris, Joyent are actually to an extent, mimicking a form of “proprietary vendor lock-in” they champion against, simply due to Open Solaris’ relatively low footprint and the extra complexity of having to learn yet another environment. (↩)

I’ve been working on something. It’ll be live soon and it’s a whole new direction. More soon.

On a side-note — filed under the special-things-that-turn-one-homicidal category — something was slightly h0rked after a recent hosting upgrade. I think I’ve nailed it now. Mostly.

“Uh, we had a slight weapons malfunction, but uh… everything’s perfectly all right now. We’re fine. We’re all fine here now, thank you. How are you?”

Worth mentioning at this point, is my natural hatred when it comes to moving Wordpress — the database is very fragile and entirely prone to crash. It’s worse when months of backups all share the same flaw that was exposed during the move and db server upgrade. Upshot?

Always have a plan D, for when A and B and C all crash and burn.

Remove the juice, wait, add juice and many-a-system-issue will disappear in a puff of imaginary smoke. Then there is the human body and mind. Which in many ways responds in the same manner. Swap juice with stress and effort, and the same results can be observed.

In a little under two weeks, I will be power-cycling. Two weeks of down-time to simply “live” without the every-day work stresses or worries will be bliss.

There are things I need to do. I need to empower myself to stop. Now, that may sound like entirely existential bull-shit but in a way it’s true. We all have the power to stop the fair-ground ride and to step away from the machine.

The hard part is realising that the ride is self managed and it’s all too easy to bring “some stuff” home to do during the break. Which makes a mockery of the entire point of stopping in the first place.

I won’t actually stop, of course. No one really does. But instead of working for “the man”, I’ll be working on personal projects.. and a great deal of effort will go into doing very little. Apart from perhaps following up on potential variables and options for a move back to New Zealand.

Which, aside from Japan, is probably one of the single most amazing places one might chose to work. Not that I have lived or worked in Japan, but I have it on good authority that it is a damn fine place to live, doubly so outside of the major centres.

Which, perhaps, is a round-a-bout way of saying that I am coming to a fork in the road, and not just in the blogging sense. Change should be embraced — the tricky part is working out just quite how to embrace it best.

The Dead Collector: Bring out yer dead.
Large Man with Dead Body: Here’s one.
The Dead Collector: That’ll be ninepence.
The Dead Body That Claims It Isn’t: I’m not dead.
The Dead Collector: What?
Large Man with Dead Body: Nothing. There’s your ninepence.
The Dead Body That Claims It Isn’t: I’m not dead.
The Dead Collector: ‘Ere, he says he’s not dead.
Large Man with Dead Body: Yes he is.
The Dead Body That Claims It Isn’t: I’m not.
The Dead Collector: He isn’t.

Life, man. I’m living it. Every day. Lately that’s meant living three days worth in each and every day.

Work pressures are such that I can barely pause for respite, without the inexorable feeling that I am somehow tumbling forward at an ever-increasing rate of knots. This is the time of the year when one simultaneously somehow manages to feel re-energised and challenged, yet entirely exhausted by the fact that that is a bloody tiring process.

It’s not that I hate that, either. I’ve achieved more in the last two weeks than I have in the last two months — at least in respect of my day job — the flip side however, is it cuts deeply into spare time; forcing other projects onto the back burner. The good news is that I have two weeks of leave coming up soon.. oh to be able to recharge and cut back to some of the projects I really want to progress. Joy.

In the mean time, it has exposed an entirely unnerving process whereby the content I read on a daily basis, some of which is likely to be your very own, ends up growing at a frightening rate. Inbox zero? Pah! 900+ unread RSS items (it will breach 1100 tomorrow) and a stack load of design work and code for a new project, both growing moss at an almost genetically engineered rate.

It seems right-now is when everything happens, yet right-now isn’t time I actually happen to have. I desperately love the universe we live in. It is truly remarkable in it’s ironic tendency to somehow “know” the exact moment that one is entirely busy and provides even more stimuli and input at that exact moment.

So whilst this may indeed be the silly season, I for one welcome our crazy-inducing overlords. There’s something to be said for smashing through the pain barrier to deliver a result no-one expects you to possibly pull off in the time-frames given.

And then I’ll enjoy the upcoming down-time, because it will seem all the more restful as a result. So.. not at all dead. Still very much alive (thank-you to those who have asked) and not the least bit, well, busy.

“Instead, I find that download options for commercial software are quite rare. Even when the download option is available, you end up paying the same price as retail or even more.”

This is a point-of view that crops up every so often, typically backed up by a small number of examples where a retail price (on sale) is less than the online option. On the face of it, that’s a pretty damning indictment of the online e-tail model, right? Surely it should be cheaper to procure a product via electronic download, rather than shrink wrapped? Of course. To a point.

Given pressing a DVD in a big run is typically counted in cents, with packaging and documentation costs substantively reduced on big runs the costs to actually manufacture a product typically pale in comparison to delivery and display costs. We pay for the convenience of being able to walk into a store and uplift a product. Why would the convenience of immediate delivery be any different? Doesn’t that have value, too?

And what is also frequently missed is that a great many products sold via the electronic model are not only available before they hit the store shelves but often launched, or pre-launched, at a reduced price. Which is totally unlike the typical retail model, where pricing starts off as high as the market will stand and then starts to drop as the distributor switches gears to try and offload as many units as possible.

Jeff picks out Steam as a good example of where price parity has taken a nose dive — because one or two titles are more than a retail sale price — title yet The Orange Box, one of the higher selling titles for Steam, can actually retail for more than the electronic download.

Don’t take my word for it, here it is in black and white. Other than Amazon — who continue their survival strategy by undercutting everyone else on principle — the pricing is at the very least on par (freight costs can cancel out online sale savings). If you take the walk-in-store approach, you will frequently pay more.

If you’re one of the many millions of potential customers not currently residing in the US of A, then digital delivery orders can result in massive savings. The very same Orange Box sold in the US for $49.95, retails for a good deal more here.

Jeff is highlighting that there are exceptions to any rule. Which is true. Just as one can pay three different prices for the same product at three different locations, online ordering may not always work out cheaper for every single instance of n at any single moment of time.

It’s not a perfect model and there can be both very good and very bad times to drop some coin. However.. the convenience, potential speed of delivery, reduction in packaging waste and early release offers, combined with frequent discounts both at-launch and on-going e-tailer sales, do still make them just as much a valid avenue as visiting the local electronic game store.

Jeff ends his article thusly:

“It seems to me that, in the area of digital distribution efficiencies, commercial software still has a lot to learn from the open source world– where everything is downloadable by design. I hope they can adapt before they’re forced into extinction.”

This is a great ethic to relate to. If we didn’t live in a world where capitalism is the order of the day. But we do. And this last comment seemingly ignores the trend where an every increasing amount of software is available and maintainable online.

It should also be noted that retailers want the electronic model to fail — and they’ll try real hard to crack pricing on a few select titles to make it look like it has. Scratch the surface though and typical higher-than-download shelf pricing belies that the old bait-and-hook retail model is still just as prevalent.

Like any purchase choice, shop around and find the deal that best suits you. Many options are infinitely preferable to none and ignoring the purchase-and-download model on the principle that there is still a price tag involved only serves to further limit options.

“The desktop background, as I see it, is completely superfluous. My desktop “background” right now is plain black. And that doesn’t bother me in the least, because none of it is visible.”

I agree with Jeff on this point. The background shouldn’t define how we get things done.

Perhaps a little show-and-tell is in order, to illustrate. This is my current home-based workspace. I don’t have too many applications actively running right now and the desktop is pretty much devoid of all but the drive icons — it was taken during the creation of this article.

For the more curious among you, the two Kanji characters represent Ninja. From time to time I enjoy cycling out desktop imagery that doesn’t have English typography. It reminds me to think outside the square and not continuously seek the comfortable, or the known, to take risks and try new things.

Back to the subject at hand. The desktop shouldn’t be a scrap yard, where tens of dozens of icons, folders and random cruft congregate waiting for something to happen, be it recycling or eventual destruction. In such a chaotic environment resources can easily be lost, or accidentally purged.

Any gains in the ’single basket’ approach to storage are lost if the basket goes missing. Rather, the desktop should act like a lobby, providing the launch point to your next destination. Jeff observes the following:

“The desktop is like an aesthetically pleasing airport we must occasionally pass through before arriving at our real destinations: a web browser, a word processor, an IDE, a graphics editor, etcetera.”

Which neatly dovetails into a piece I wrote recently, regarding experimenting with minimalism and plain colour backgrounds. Although Jeff mentions that he has “browser windows and programs — the things I’m actually doing — covering all three monitors..” I find that keeping applications minimised when not in use frees up resources for those applications that you are using, improving the response times and resulting efficiencies.

Given Jeff has three monitors in action, the numbers of open applications that he can run concurrently without overcrowding is obviously greater than if one uses one, or even two monitors. In my case, I have a wide-screen display at home and use two at work.

When you take a break¹ save your work² and minimise any focused applications. Those two actions help improve the break quality as you won’t be tempted to worry about any current activity to the same degree. It’s saved and stowed away safely.

When you return, you’ll be greeted with a tidy workspace and a system ready to respond to your needs, not as a user forced to bend to the will of the system. I have written more in the last three days alone than in the last two weeks. Much of that can be attributed to a renewed sense of achievement that a better work space can provide.

I’ve found increased enthusiasm in using the tools I presently have and, well, I feel as though I’m getting more done, without any additional stress. And there really is nothing quite like sitting down to a system just as refreshed as it’s user, that isn’t overburdened with several applications all vying for attention and is ready to spring into action.

1. a regular stretch or beverage run should be a regular activity (↩)
2. don’t be tempted to think “just one more sentence..” (↩)

“The operating system is dead. Long live the operating system. If the future is network applications, the past is the operating system.”

I think David has oversimplified the concept of an OS-less and server-less environment just a tad. Hardware and software will still need basic building blocks able to run the software of the future, be it server-client or de-centralised. He also notes:

A Web operating system will not emerge. I’m sure this point will be argued vigorously. But I’m right.

That is a give-in. Because the web isn't defined by being an "operating system", rather it is an increasingly interactive environment. So it's hardly breaking new ground to suggest "the internet" won't be an operating system, as it were. David also states:

Because the operating system is pushed down the stack, doesn’t mean what replaces it is an operating system.

To suggest a less complex OS isn't really an OS at all, side-steps the reality than an Operating System is defined primarily by it's function, not how big, or small it is. Or which parts of it reside where. This, again, is a questions of semantics.

A duck doesn't suddenly become an elk the moment it gains a slightly more efficient plumage and flies faster. It is still a duck. Just because many operating systems are moving towards a decentralised nirvana, doesn't suddenly mean they stop existing entirely.

As an example, the software that allows the iPhone to surf the web, access email and receive calls all hinges on a base install of OS X. It may be hidden, indeed one may potentially never actually see it, but it's still there.

Joyent rely heavily on Solaris and large server farms. Without either they would be little more than an idea looking for some way to find fruition. Indeed without that standardised, documented operating system they would never have been able to build their various products to begin with. It seems at odds with sense to be so quick to claim the imminent death of the very thing that helps gives one’s business life.

Further reading, however, illustrates that David contradicts and concedes some points:

This doesn’t mean a client edge device won’t have an operating system. It just won’t matter. Changes have been afoot to usher in this new era. The web is one.

So we’ve gone from no operating system, web or local, to an invisible one that doesn’t matter. I don’t typically “see” oxygen, but I still require it to live. Future technologies will be just the same, in that they too will require a basic operating system or environment to function. That may be entirely built into the device itself, or may be some kind of downloaded “intelligent code”. It may be a very different beast, but it will still be there.

This is really so much more about de-decentralised, portable and application focused computing and evolution of technology, both at the user and producer levels, not the death of the operating system itself. Smart devices are cropping up everywhere and David is right, computing as we know it will change massively in the future. But in order that those new devices can pass information and work in a semi-predictable way, they will still require a common tongue and portable code to co-exist.

How they will be constructed, where they will reside and what remnants of the current model will remain is something that will challenge us for many years to come.

“I'm not against cross-platform libraries, or cross-platform applications. But giving the same UI to every OS isn't a good solution.”

It isn't a good solution, but one often mooted as the 'best' compromise. The downside of that middle ground choice is that it will offset any possible advantages gained by presenting an interface that does not fit it's environment.

Good design separates presentation from engine; web standards are built on that very concept.

If you use OS X, you're used to a certain user interface, just as a Windows or perhaps Gnome-on-Linux user is familiar with theirs. The GUI should be predictable across applications and make a certain logical sense. Cross platform UI breaks that fundimental design concept by introducing a foreign element, which whilst keeping strong brand integrity, fails to keep a reasonable standard of UI integration in return.

Mozilla developers working on Firefox 3 have begun to understand that embracing each Operating System's design does not mandate that the ‘underware’ be different, rather that the clothes Firefox wears under each match the surrounding environment.

“Visual integration with Windows and OS X is our primary objective for the Firefox 3 refresh.”

Of course, views differ and Simen of dailymeh responds:

“All else often isn’t equal, because it’s often a choice between a stripped-down version or no version at all and a slightly sub-optimal solution.”

Which is a very valid point, but falls flat when the tools already exist to make cross platform code happen without enforcing the need for a constant user interface as well. The days of settling for second best or none at all have long since passed. Web based technologies have certainly driven that change, but ‘fat’ application development tools are catching up, fast.

All the brand identity in the world equates to exactly zero if absolutely no-one uses your product due to a confused and unusable cross-platform interface. We’re no longer bound to a narrow range of applications so the old rules of “it’s our way or the highway” don’t fit with the modern world.

Perhaps that is why so many developers continue turning to web based application technologies. Because it does allow for very strong branding yet will still look, at a fundamental level, virtually the same across any number of platforms. The latest Mozilla lab product, Prism, jettisons all but the most rudimentary user interface, leveraging the existing web navigation instead. And that concept is raising eyebrows.

It’s just not the done thing, throwing away the standardised cross platform local user interface. Yet in ever increasing web application cases the local interface, short of basic core controls, becomes entirely redundant.

The future of application development is increasingly cross platform in nature, of that there is no contest. Whether or not that is a wise idea is also no longer a relevant question. In fact I’m all for it — diversity is a strong survival technique and allows for a wider range of tools in return. But for off-line applications to succeed, they will increasingly face direct competition from web based technologies, and experience the exact same growing pains.

Part of that process is separating engine from user interface and treating the two halves, not just the whole. The promising sign is that that is already happening.

It does appear some cooler heads have surfaced with a little voice of reason regarding the decision by Apple to not roll Sun Java 1.6 into the first release of Leopard. Ben Galbraith, co-founder of Ajaxian, has stepped up and writes:

"So, what, there are some bugs in OS X 10.5.0? They didn't have 1.6 ready out-of-the-chutes? Big deal. Give it time, just like we have with every release of Java since OS X first shipped with 10.3."

Ben also points out a working solution for anyone looking to develop that has made noises about shifting platforms:

".. now, thanks to Parallels/Fusion, we actually have a great alternative for the impatient."

It's not exactly breaking new ground using a virtual environment for development, indeed some might suggest it's the ideal approach. So as Gruber has pointed out, this is little more than a storm in a teacup.

He's not alone in that view either as Eric Burke points out:

"Panic! Panic! No, wait. Let’s learn from history. I spent some time this afternoon putting together [a] timeline comparing Sun’s Java releases with Apple’s Java release.."

The timeline clearly shows that Java updates have universally occurred within short order of a major milestone release. Certainly for both Panther and Tiger. This isn't a new thing. It's also not just an Apple thing. Unlike our solar system's celestial bodies, most Operating Systems do not revolve around Sun.

From my point-of-view, Java may well be one of 'the' platforms of choice for the mobile space, but it's place is far less cemented on the desktop. Where Leopard is in use. I seldom see Java outside of the web browser, be it in Windows, OS X or even Linux; indeed aside from a small handful of applications, few actually will.