1920 x 1200

I nicknamed this fractal “wheat-field” -  and if you look at the full size image, it isn’t hard to tell why.  Wheat-fields, aside from reminding me of long, boring drives across the prairies, remind me of my early teens.  I spent a decent chunk of my summers in those years visiting one uncle or another out in the countryside, and I can’t say I have any bad memories of that.  As I got older, I found those long boring drives were a lot easier to take than some of the people I made the drive with, because the prairies hold a lot of good memories for me.

The funny thing about memories, is that they tend to magnify or diminish with time.  Depending on your personality and outlook, the good memories magnify, and the bad ones diminish – or the other way around.  Most days, I find that the older the memory, the more likely it is to be a good one.

So it might not surprise you that when I revisit a TV show or movie from my youth, sometimes they just don’t measure up to the memory.  I’m sure you’ve all experienced the same syndrome.

But sometimes…  Well, it is as good as you remember.  Maybe even better.

I recently stumbled across an episode of an old TV show from those early teen years, Amazing Stories.  It was a short lived series, mostly 1/2 hour episodes, with a few 1 hour specials.  In retrospect, it was doomed to failure in the TV world.  It consisted of set piece stories, in the vein of Twilight Zone, or Outer Limits, so people didn’t have any kind of ongoing characters or story line to grab hold of.  Add to that it was written and directed by Steven Spielberg, who had a habit of placing quality as a higher priority than maintaining a TV appropriate budget…  Well, you can pretty much figure out where that would lead.  I’m somewhat surprised it lasted 2 seasons.

One episode in particular really stood out for me.   It may sound weird, but it was an episode I only ever saw the once, and over the years I’d think of it every now and then at the strangest times.  Just sort of pop into my head for no reason.

I finally got a chance to watch it again, and I have to say, it was every bit as good as I remember.  Maybe even better.  It was an engaging, well written, high production value episode.  And the cast.  That was a part I didn’t remember being so…  high powered.  In one of those rare glimpses of a number of actors, all in the same little known show, “before they were famous.”  Kevin Costner (in his pre-arrogant douche could still act kinda days), Casey Siemaszko,  Kiefer Sutherland, and Anthony LaPaglia.

Directed by Steven Spielberg.

In a little known episode of a short run TV series that most of you have probably never heard of.

But if you did hear of it, and remember seeing it, I can assure you that yes, it was as good as you remembered.

1920 x 1200

I colour shifted this fractal a bit.  It reminds me of a slightly distorted Yin without the accompanying Yang.  (Or is it Yang without the Yin?)

I was flipping through my feeds, and came across an interesting bit of a rant against Facebook and they’re poor security implementation on a security research blog.  I found it funny, in a technical geek philosomifizing kind of way.  I’m guessing that most of you who click on that link might not get the joke. (Hint – The joke is about a security researcher who actually uses Facebook, then complains about the security – talk about shooting fish in a barrel).  Sure, I use Facebook, but I use it with the assumption that the whole platform is fundamentally insecure.  That’s sort of a given.

But it ties in with with something I read earlier in the day about Cussing in Code Comments.  The article itself was about which programming language tends to have the most swears embedded in the comments, but it pointed to one specific code comment that I’m sure has ended up, in one form or another, in thousands of pieces of software over the years:  “Fuck it.  Let’s Release.”

It speaks to an industry wide problem.  Sooner or later, you run out of time to fix things, and you have to suck it up and release something that may not be perfect, but it works.  Kinda sorta.

If you’re not in the industry, or are a “certain type” of manager in the industry, you might look at that kind of attitude with disdain – can’t people take pride in their work anymore?  But if you’ve ever spent time generating code, even if its a simple BASH script or BATCH file, I can almost guarantee that you’ve said something similar to yourself, even if you haven’t actually embedded the comment into the code.

Essentially, it comes down to 2 key issues.

First: This stuff is complicated.  No, really, software is way more complicated than you realize.  This “simple” little blog is a prime example of how complicated software has become.  This site uses WordPress as its prime engine, with some add-ons and minor modifications I’ve done myself.  All told, the code accounts for over 2,000 separate files that interact with each other, some larger than others.  At a guess, I would put the word count in the code to somewhere similar to twice that of The Lord of the Rings.  It represents the accumulated efforts of several hundred people.  And for all that code to work, it has to interact in complex ways with a Web Sever and Database software.  And then for you to see it, the generated output has to be interpreted by your web browser – one of dozens of differnt browsers running on a number of different platforms, and each browser and platform combination will look at the output in slightly different ways.  The fact that it works at all is somewhat amazing.  Even there, I actually used certain “edit hacks” when I’m composing a new post, just to make sure things line up reasonably well in all those different browsers.

And in modern code terms, this is genuinely a fairly simple application.

Second: This is a deadline driven, “publish or perish” industry.  With an infinite amount of time, and an infinite amount of resources, it might just be possible to write the “perfect” software.  But that just isn’t the world we live in.  This is an intensely competitive industry, and the advantage goes to the first person/company to release a product for a given sector.  Sure, things can get disrupted by an outstanding newcomer who comes up with an innovative way of doing the same old thing, but that’s the exception to the rule.  Look at Microsoft Office.  I will concede that MS Office is a pretty amazing package, it does a lot of different things very well, and in a way that’s intuitive for most people.  At the same time, it’s bloated, full of bugs, and far more expensive than it needs to be.  It maintains market share not because there aren’t excellent alternatives, but because it already HAS market share.  MS Office is the de facto standard in offices around the world.  Whatever the problems and costs associated with the software, businesses use it because they know that they are less likely to have to train their staff how to use it, and that the documents they generate with it are virtually guaranteed to be readable by the widest possible audience.

That doesn’t mean that Microsoft hasn’t kept innovating the product, evolving it and squashing bugs – they have to or someone else will come up with something with significant enough improvements that it will warrant businesses to change what they use.  It isn’t like others aren’t trying to take the market away from them.

This puts enormous pressure on software companies to publish quickly.  The modern motto is “Release Early, Release Often” – a philosophy that came out of the open source movement, and has since permeated the industry.  The roots of this type of engineering go back even further, at least to the 50′s, but it really grabbed hold of the software industry in the late 90s onward.

The end result is that developers, and to an extent IT administrators in general, are under some intense deadline pressure.   And the economics of it mean that you can only have so many people working on a given problem.  Limited resources, limited time.  Make Rocket Go NOW!

So mistakes happen.  There isn’t enough time to test all the possible scenarios.  Everyone will do what they can, test what they can.  Eventually, though, there just won’t be enough time to make it perfect.  So coders will continue to swear, and sooner or later, everyone ends up saying “Fuck it, let’s release.”

1920 x 1200 - Click for Full Size

I took a couple of days off from posting, so let’s get back into the swing of things with some vague thought waving.

I spend far too much time reading tech-centric news and forums.  And one of the things that strikes me, is how often the tech pundits, professional journalists, and industry insiders so completely miss the mark in so many ways.  Let’s take a look at a couple of recent examples of how almost everyone got it wrong, in ways that directly affect both the industry, and the consumer.

First off, the NetBook revolution.  A few years ago when Asus released the first iterations of the “EEE PC”, which can largely be viewed as the precedent for all NetBooks to follow, it was widely and almost universally derided.  It was seen as, at best, an underpowered “niche” product with no real future.  The screen was too small, the keyboard almost unworkable, and in terms of processing power, it couldn’t handle much beyond basic web surfing, email, and word processing.  Steve Jobs famously quipped “We don’t know how to make a $500 computer that’s not a piece of junk…” In short, they viewed it as a trinket, a toy.

And then sales took off.

So how did they get it so very wrong?

Because the pundits and industry insiders aren’t the average user, and they don’t use computers like the average user.  Often, they’re just a bit too savvy for their own good.  In this case, they completely missed the point of the device and the entire category.  For these people, and people like me, we’re constantly looking at how to get MORE from our systems.  We need them to edit video, handle complex software, and to do it quickly.  What we all failed to see was that for a lot of people, a simple, small and lightweight device that could handle a few basic tasks was actually incredibly useful.  It was something you could take with you anywhere to accomplish those things that most people actually spend 90% or more of their computer time doing – basic web surfing, handling a few emails, and maybe working on the odd spreadsheet or document.  And put that into an attractive price point (under $500 – and for as little as $300), and Asus was suddenly able to fill a niche that no one knew was even there.  It transformed the market.  And that transformation lead to another great innovation that no one saw coming.

When the iPad was announced, it was met with a great deal of skepticism, to say the least.  What’s the point?  An over-sized iPhone that still can’t play Flash?  Not to mention, it wasn’t exactly as revolutionary as Apple (and Steve Jobs) would have us believe.  Tablets had been around for over a decade, by this point, and they had simply never taken off.  They were a niche product, useful for very specific purposes in certain industries.

But they weren’t the iPad, and what was revolutionary about the iPad wasn’t the hardware.  It was how Apple put the hardware to use.  All previous tablet computers had essentially been a keyboard-less laptop with a touch screen, with an operating system on it that had never been designed to be used without a keyboard and a mouse.  They were kludgy, difficult to use, and they also tended to be outrageously expensive, usually on the order to 1.5 to 2 times as expensive as a similarly powered laptop.

The iPad wasn’t released with a desktop operating system.  If it had been released with a “customized” version of OSX, (or if some other company had released it with Windows or Linux), it would have failed in much the same way as all previous tablet computers.  The iPad came out with a slightly tweaked new version of IOS, the iPhone operating system that was designed from the ground up to be used on a touch screen, and had been tweaked and refined after being in actual consumer’s hands for a couple of years.  And it came out at an attractive price.  Instead of pricing the device in line with other tablets, they priced it in line with what it really was: an over-sized smart phone.

And again, it flew off the shelves, leaving competitors to come up with competing products in an ad-hoc manner, and they still haven’t caught up yet.

In both cases, the pundits, analysts, and insiders missed the point for the same reason.  They (or should I say we, because I was as wrong in both cases as so many others), were wrong because we were all entrenched in our way of working.  So intimately immersed in the details of individual trees that we completely failed to see the Forest as a much larger, more complex and diverse ecosystem.

And it can be very difficult to break out of that myopia.

So the next time you read an article about someone predicting “The Next Big Thing”, the “Imminent Demise” of a technology, or the “Destined Failure” of a new product class, take a step back, and remember who’s making the prediction.  And then try and look at the technology or device as it is, how it relates to you, and whether it might make a difference to you.  Then come to your own conclusions.  To be honest, an outsider has as much a chance of being right (or wrong) as so many experts.

1920 x 1200 - Click for Full Size

I really, really, like this fractal.  It has a beautiful colour palette, and the texture resembles brush strokes.

Normally, it would be really easy to get me going on a rant about the pointlessness of modern art.  Randomly slapping a paint brush against a canvas, throwing paint at a wall sized canvas, etc., to me, just isn’t art.  The only art involved is in convincing someone to pay a lot of money for it.

But, maybe, sometimes, modern art is more than random paint splashes and hucksterism.  Don’t get me wrong, slapping a red stripe down a blue painted canvas shouldn’t be worth $1.8 million, especially if the bill is being footed by taxpayers.  On the other hand, sometimes modern art can be thought provoking, or at least pleasant to look at.

1920 x 1200 - Click for Full Size

So, I started messing around with some new (to me) fractal software, XaoS, on my Linux box, and this is the first one I’ve rendered out and captured.

Like a lot of things in Linux land, the UI for this software leaves a lot to be desired, and there’s a bit more of a learning curve for it than there is with equivalent software on proprietary platforms.  In fact, if you read through their project pages, they indicate that the version they’ve released for Mac has a much better UI.

But hey, to use that, I’d have to buy a Mac, which should happen sometime around the heat death of the universe.

On the other hand, like a lot of things in Linux land, once you get past the dodgy User Interface, you actually have a lot more fine grained control over the software than you tend to in the equivalents on  Windows or Mac.  Live zoom, instant capturing, really good control over the algo, colour…  And it’s really light on system resources.  It renders the fractals out instantaneously on my Linux box, which has much lower hardware specs than my Windows box.

I had a lot of fun coming up with this one.  It’s a zoom of an Octal Mandelbrot set fractal, with some colour tweaks.  I then took the resulting capture, and did some colour, hue, and level balancing in GIMP, and this is the end result.  Long story short, this is by far the most “manipulated” fractal I’ve posted here.

Oh, and for those of you who are about to jump up and down and complain about GIMP (Gnu Image Manipulation Program), let me just say the following:

• It’s about $700 cheaper than Photoshop (as in, GIMP is free, and I just checked the price for Photoshop at the Adobe site).
• You probably haven’t used it lately – the newest versions of GIMP are actually quite easy to use and fairly feature packed.
• It will run/look the same on Windows, Mac, or Linux.  For me, the Linux part is kinda important.
• It’s incredibly lightweight.  The latest versions of Photoshop are fat bloated resource hogs.
• All those fancy features in Photoshop?  99% of you will only use a tiny fraction of them, and the equivalent features for those are in GIMP.  For the other 1% of you, there’s a plugin available.  In fact the fancy liquid rescaling that was recently released in Photoshop was available as a GIMP plugin for a year and a half  before it was released as “ground breaking feature” for Photoshop. (If you don’t know what liquid rescaling is, it’s way too hard for me to describe – search YouTube for “gimp liquid rescale” or just go to the plugin page for an idea – it is a vastly cool thing to be able to do to an image).
• My wife, the one with the BFA in Graphic Arts, who currently works in architectural design and drafting, abandoned Photoshop for GIMP several years ago and doesn’t for a minute regret the decision.  She actually finds that all the key features that you use on a regular basis are easier to find in GIMP than in Photoshop, and she’s the closest thing I have on hand to a graphic arts expert…  So it’s nice that she IS a graphic arts expert.

So, for a $700 savings, (which equates, I remind you, to FREE) you get a highly workable graphics editor, that even if you’re a pro and need all the “deep” features, you can get plugins and extensions from a pretty vast repository.

FOSS is a pretty fun world, and it keeps getting better by the day.

1920 x 1200

I know a few people who are so scared of pretty much everything, that it takes an intense act of fortitude to even leave the house. They worry about muggings, terrorism, the weather, disease (which often, weirdly and irrationally, goes hand in hand with fear of vaccinations), children getting hung up in drop rail cribs, “sudden acceleration syndrome” in their cars, meteor strikes… You name it, and they’re afraid of it.

And it’s just so pointless, not to mention out of step with what’s actually going on in the world, that I’m not sure whether to pity them, laugh at their ignorance, or rage against those who have a vested interest in perpetuating it.

I can’t possibly list all the things people fear, and some of them pretty much warrant their own posts (fear of G.M. foods, fear of Nuke Power, I’m eyeballing you, you groundless, media stoked, irrational paranoias), but I think maybe I can tackle a couple of them.  Hey, maybe one person out there will actually pay attention and fall under a sudden grip of rationalism.  Don’t laugh! It could happen!

We’ll start with one of my favourites, the fear of terrorism.

And right there, my mind almost froze contemplating the irrationality of it.  Do you have any idea how vanishingly small the odds are of being caught in act of terrorism? Or that terrorism peaked globally in the late 80s – early 90s, and in spite of rising, ever so slightly since 9-11 (source), are roughly half now what they were at 88-91 peak? Here’s some fun numbers:

• Lifetime risk of being killed in a car accident, 1 in 83.
• Lifetime odds of being killed crossing the street (for crying out loud), 1 in 625
• Lifetime odds of drowning, 1 in 1100
• If the global Jihad suddenly got it’s act together and was able to pull off a 9-11 scale attack at least once a year (on top of the normal background noise of terrorism that mostly happens in the Middle East), 1 in 1300
• (source)

That’s right, even given a worst case scenario of terrorism, you’re more than twice as likely to be killed crossing the street as you are to be killed by a terrorist, and fifteen times as likely to die in a car accident.  Don’t believe me?  How many people have you known who have been killed in a car wreck?  How many people have you known who’ve been killed by a terrorist?  For most people outside of New York, the answer to the first question will probably be at least one or two.  To the other question, none.

So here’s a thought: Ditch the full body scanners at the airport, and instead stop yakking on your cel phone while you’re walking/driving around downtown.

Next up: Fear of crime.  I don’t have any good statistics for my country, so let’s take our most beloved wretched hive of scum and villainy (the United States) as a good baseline. In the past twenty years, violent crime in the US has dropped by half, property crime is down by 40%,  but our perception of crime is at an all time high (source).  How the heck could that be?

A big part of it is the 24 hour news cycle.  There’s an old adage in journalism: “If it bleeds, it leads.”  Violence, and violent crimes, are good for ratings.  So the news networks go out of their way to seek out crime stories, splash them on the screen, and milk the crap out of it until the ratings start to slide, then go find something else.  And this plays against our very basic instincts to be afraid of that movement in the shadows.

Also, “getting tough on crime” and a good “war on terrorism”, are really good platforms for getting politicians elected.  Much more effective than the “Elect me, because things are pretty good right now, so I don’t plan on doing anything” election campaign.

And I could go into all the other fears you might have, but really, it would just be an exercise in throwing statistics at the wall.  Really, the only statistic that matters is that, on average, you will live longer now than at any point in recorded history.

So you might as well get out there and enjoy it, because you’re going to be around a lot longer than you might think.

1920 x 1200

Kinda neat one…  Not often you end up rendering a fractal that ends up being all straight lines.  At least, not with the software/methods I use.

So, it’s Saturday night, and here I sit locked down in front of the keyboard…  Again.  Like so many Saturday nightsw before me, and so many after.  I’m really just taking a brief sanity break, and then will be back at it until the wee hours.

I was hard at it until about 9am this morning…  Slept a bit, spent some time with the family, then it was time to work again.  This is an unusually busy weekend for me – a whole raft of problems conspired to make me very busy with problems that, because of the SLA (Service Level Agreement)
we have with our customers, need to get solved so I can hand off 150 imaged, live on the web, servers by noon on Sunday.

I could complain…  Actually, I do complain about it.  But really it’s an industry wide problem.  Once you pass a certain level, there’s a long stretch in your career where your hours are pretty much 24/7.  That doesn’t mean I work all day everyday.  It just means I’m on the hook should something go wrong.  The issue is that there’s a whole lot of systems that I can troubleshoot quicker and easier than most other people in my company.

That doesn’t mean I’m smarter, or even more knowledgeable than they are.  It just means there’s a big piece of our infrastructure that I’m more familiar with.  Other parts, there are other guys who are more familiar with it, and they’re on the hook for the same kind of hours as I am.

Ideally there should be “knowledge redundancy” across the board.  And to a certain extent there is.  If I get hit by a bus tomorrow, our company will survive just fine.  Someone else will shift priorities, start working on the same chunk of our infrastructure, and working from a good general base of familiarity with IT in general and how our company manages systems in specific, over time will gain the same level of intimate knowledge of the bit of infrastructure that’s currently “my problem”.

The problem is really twofold.

1.  Budgetary/traditional staffing standards.  Companies have gotten use to having a key “Unix guy” or “Windows server guy” or “Storage appliance guy” or whatever…  A key individual who gets to know a specific set of systems.  Having a lead tech for a given area makes sense, because we don’t come cheap.  Usually, this person will have 1 or more people working under them (right now I’m blessed to have two PFY‘s), but there is usually a steep knowledge gap between the two.  With the day to day pressures of business, the PFY’s usually end up doing the grunt/chore work that’s not time sensitive, and the BOFH will end up perpetually on call to deal with the bigger, more time sensitive problems.

Here’s the funny thing, there’s actually a deeper hierarchy to this.  While I may be the BOFH to two PFY’s, I’m also the PFY to a completely different BOFH.  Which is fairly standard in really large IT infrastructures.  At some point, you might pull out of that loop and move into management or architecture design, but a lot of people spend their entire IT careers in the PFY/BOFH hierarchy loop.  Probably most people.  And if that’s the case, then you’re almost never not on call, and the hours suck.

2. Personality.  It takes a certain type of obsessive compulsive person to become any good in virtually any part of IT, and when I’m being honest with myself, I take a bit of perverse pride in the ridiculous hours I’ll put in to solve niggling little problems that most people could care less about.  And that just adds to the hours.

Which brings us to the title of the post.  Sure, I make reasonably good money.  Heck, I’d cleared my mortgage when I was 35.  But there’s a price to be paid for that.  And that price is time.  I let myself dig too deeply into things, put in foolishly long hours, and it takes a toll.  Emotionally, physically, and on my family life.

And I find myself thinking about this more and more as my daughter grows.  Do we really want to raise her to be a nerdling techie?  I dunno.  I suspect she might be happier doing something else.

1920 x 1200

This is kind of a neat fractal.  Reminds me of stirring your coffee before you drip in the cream.  Which is appropriate, because one of the scientific uses for fractals is to simulate fluid dynamics.  This image might give you a sense of just how that works.

It also provides a good segue into my “thought for the day” – which is that technology is much more impressive when you have an understanding of what it’s actually doing.

For a lot of people out there, probably most people, computers are nothing but a magic mystery box.  They have no understanding of what’s going on under the hood, and quite often don’t want to understand.  Which is a shame, really, because if you have even a glimmer of an understanding, then they become just that much more amazing.

I’m thinking about this because I just got my wife a spiffy new smart-phone.  In itself, it’s a marvel.  It’s more powerful, and has better screen resolution, than my first laptop did back in the 90s.  Much, much more powerful.  But it’s a bit hard to wrap your head around all the things that a modern smart-phone can do, so let’s step away from that, but stick with me, it all tracks together.

Now, I have a beautiful little two year old daughter.  As with all beautiful little two year olds, she can get a bit cranky when she’s getting dragged along with us when we have to do “adult” things.  A little trick I’ve used is to have a stack of short, 7 minute videos loaded onto my own phone, that I can quickly bring up and show her.

My wife’s phone is a different brand altogether, with different resolution etc. than mine, so I can’t just copy the videos from my phone over to hers.  But I have all the originals on my computer, so it’s really not that big a deal to convert the originals and load them onto my wife’s phone.  And that, my friends, is amazing – if you understand what’s really going on under the hood. (p.s. I live in a country where “format shifting” videos is perfectly legal – this is not the case everywhere, so mind your local copyright laws).

I’ve got the process pretty automated, but let’s explain, for a second, what’s actually going on.  It needs to change the format of the video, and resize it, and re-sync the audio track. No biggie right?

Ok, here’s an analogy for you.  Take an oil painting, re-paint it using acrylics, to perfect three quarter scale, maintaining the same balance of colours and aspect ratio.  Now do that 40 times per second, while binding the results together into a flip book and reciting the narration, all at the same time.

That’s right, each frame of the video is a picture that needs to be re-sized, have it’s format changed, and the audio track synced to it while it’s being tied together in a contiguous whole.  My 4 year old “secondary” computer is pumping through all that work at a rate of 40 frames per second.  That’s not bragging about the power of that computer – that’s just showing what even relatively older computers can do.  A modern equivalent (keeping Moore’s Law in mind that computers double in power roughly every two years), computer would be able to do this at a rate of 160 frames per second.

Now you know something.  And doesn’t that make the Magic Box seem just that much more awesome?

1920 x 1200

This fractal has both a neat pattern to it, and a fairly subtle blue colour scheme.  It actually makes a really good desktop background because the blue/black contrast allows for really visible icons on icon text.  Keeps it easy to find all your stuff.

Now, as for the title of the post…

I got thinking today about recursive feedback loops thanks to a couple of lost hours trying to figure out why a NetApp Filer was misbehaving in a way that led to an “interesting” series of phone calls between our largest customer and various levels of our own management.  After the dust had settled, I started image surfing satellite imagery, because, well, I find it soothing.

I came across the following picture, and being a tad OCD about these things, I tracked down the source on the nasa.gov site.

It’s a pretty famous shot, and it got me thinking about the state of affairs in Africa. Basically, poverty begets poor government begets lack of meaningful development begets poverty.  Repeat.

1920x1200

So, I think I’ll put the fractals at the top left of the post, so people can get to them quickly without having to plow through whatever I want to rant about on any given day.  I kinda like this one, it’s a really pleasant soft orange pattern.

But onto other stuff…

A friend, Xepol, made some good comments on my “All OS’s suck… Even Linux” post from a while back. I’m not going to say that I agree with everything he said (although some of it, I do agree with), just that he made some valid points.

I think the big thing, for the “Average Joe” to remember…  And probably even more so for the tech geeks who really get wrapped up in the debates around various OS’s, is that the three major operating systems out there have their own strengths and weaknesses.  Each one has it’s place, and it’s target user.  And within those bounds, the three major OS’s are all wondrous marvels.

Windows: We’ll start here because it’s ubiquitous.  And in a way, that’s part of it’s strength.  Odds are, if you have a Windows machine at home, you know it will run all the software you run at work.  Also, it will run all the software that your friends use.  It will work with any hardware you get from the computer store down the street.  Manufacturers design hardware with Windows in mind because, well, they’d be crazy not to.  As a manufacturer, you just don’t ignore 85-90% of the market.

Installing new software is relatively easy, and when it isn’t, it usually isn’t the fault of Windows.  Microsoft has gone to great lengths to make it simple for outside vendors to create easy to install software – so if it isn’t easy, it’s generally the fault of the software maker.

And whatever your complaints about the user interface may be, admit it…  It really didn’t take you that long to figure it out.  Keep in mind that, on a fundamental level, computers are outrageously complex, multi-use devices.  For the majority of people, Windows just isn’t that hard to figure out.

Mac/OSX: This one is actually a bit harder for me to talk about, because out of the three major environments, it’s the one I’m least familiar with.  But I do know a few of it’s strong points.

Firstly, it has an absolutely gorgeous User Interface.  Really.  It’s just really damn pretty, and quite simple to understand.  And aesthetics matter.  If you’re going to spend a lot of time in front of a monitor (as more and more of us do), it’s really nice for the interface to be nice to look at.  It just makes the whole experience that more pleasant.

And the UI is extremely well thought out, down to some very tiny details.

Also, the fact that it’s a vertically integrated environment means you can get some levels of reliability and performance that just aren’t possible with Windows (and maybe, I’ll admit, even Linux).  The same company designs the core hardware platform as designs the operating system.  And they design the two to mesh together very well indeed.  There’s a lot to be said for that.  And the hardware itself is well thought out, eye catching, with really high quality control standards.

The end result is you get a tightly controlled operating system, on a tightly controlled platform, all designed to work very well together.  And it does.

Of course, there are downsides to all this, but that’s not what I’m writing about today.

Linux: Ahh, Linux.  The OS of the mad scientist and uber nerd (sarcasm intended).  You can list Linux’s key strengths in three words.  Security.  Reliability.  Malleability.

Security: A well set up *nix box is truly a tough nut to crack.  There have been some highly misleading reports over the past few years about more holes being found in Linux on any given year than in either Windows or Mac OSX, but you have to understand the nature of how these holes are found (and fixed) to really get a grasp on how secure the system really is.  Linux is “Open Source” – which means that anyone who wants to can look right at the pre-compiled source code, and even edit it if they want to.  As a result, on any given day, there are far more people looking at the guts of Linux than any other operating system on the market, and most of these people are looking at it’s guts because they’re working with it.  In practice, this means that the person who discovers the vulnerability, is often the same person (or part of the team), that patches it.  So the vulnerability never really makes it out into “the wild” in the first place.

Reliability: Linux can be a pain to set up.  I’ll give you that.  But once you set it up and get it running…  It just runs.  And runs.  And keeps running. An example: About six months ago we retired an old server after we’d moved it’s functions onto a different physical box.  For shits and giggles, we printed out “Top” for the server, which listed a bunch of different stats, including total accumulated runtime – the length of time since the system had last been re-booted. It came in at 847 days.  Think about it.  That machine had run, without needing a reboot, for two and a half years straight.  When you look at High Availability infrastructures, you will invariably see Linux machines at their core.  As servers, their robustness is legendary.  The backbone of the Web is essentially Linux.  The latest stats show that 57.6% of active servers across the web run Linux on the back end.  If you include Google, nginx, and lighttpd (which are all Linux derivatives), then the total is 78.2% – almost as dominant in the server space as Windows is on the desktop.  It’s an often overlooked, but truly staggering statistic.  There’s a reason for this.  And that reason is reliability and scalability.  Cost, whatever some might argue, doesn’t actually enter into the equation here.  Downtime is more expensive than any difference in licensing fees for the operating system.

Malleability: Which is a fancy way of saying you can bend it to your will, reshape it, remake it, call it your own.  Because it’s open source, anyone can reshape Linux and customize it, in very fundamental ways, that just aren’t possible with any other distribution.  Don’t need a Graphical Desktop?  Then get rid of it – most distributions allow an install without it.  Only need a small chunk of it to do some very dedicated tasks?  No problem.  Feeling the need to sex it up with all kinds of desktop effects of dubious value? Go ahead (for other videos, search for “Compiz Fusion” on YouTube).Want one custom built for specific industries or consumers? Here’s a quick list of 10.  One page lists over 600 customized versions, and it doesn’t look all that up to date.

There’s a downside to all this “fragmentation”, as some would call it, but it’s also very useful.

So what the heck is my point? I still stand by my previous post, All OS’s Suck.  They all have their limitations and weak points.  At the same time, they’re all amazing, in their own way.  Just keep in mind the following questions the next time you want to decide on Operating System:

1. What do I need it to do most?
2. Which operating system does THAT the best?