Tag Archives: open source

Roboto and the open source red herring

I largely agree with John Gruber and many others than Roboto is a bit of an ungainly beast of a font, although it’s much better than the hideous thing it replaces. But I think that John is missing the mark in this statement:

This idea that designers who favor iOS criticize Android for being poorly designed just because it’s from an Apple competitor is nonsense — a bogeyman construct dreamed up by open source zealots who refuse to believe over a decade of evidence that open source UIs tend to be ugly, and that ugly UIs tend to be unpopular.

Being open source has nothing to do with it. Like almost everything in Android, Roboto is designed, used and built at the instigation of Google: it’s not like Roboto was created by an amateur font creator sat in a basement who wanted to contribute to a project.

Android’s design deficiencies have nothing to do with the source being open. Android’s design deficiencies are down to Google not being great at designing consumer products. Android could be completely closed, and it would still look the same.

Enhanced by Zemanta

On Google’s “openness”

If you want to start a flame war, post something about whether Google is truly “open” or not. Nothing in the world of technology – not even comments on the state of Steve Jobs’ health – is more likely to get people shouting at each other.

Amongst sections of the Mac community, Google gets a lot of stick over its openness. Some criticise Google for putting “open” above “usable”. Others claim that openness is nothing more than a marketing bullet point for Google, and point to its failure to release source code for Honeycomb or its total silence over the core algorithms that power search and ads.

I don’t think Google’s openness is “just” a way to mislead – I genuinely think that internally, there’s a lot of commitment to being as open as is commensurate with being a profitable company.

Some of their efforts are extremely valuable: for example, while I think WebM is crapola, it’s valuable to have a freely-licensable codec that will (hopefully) be widely supported. I doubt that MPEG-LA would have been as generous with the terms for H.264 as they are currently had Google not waved the big stick. And that’s an area where there’s little direct revenue implication for Google.

Having said that, it’s clear that at some point internally, the idea got floated that “we are open” was a good marketing point, and that’s where things began to go wrong. It’s almost impossible for a company which creates code, delivers online services, or (for that matter) makes hardware to be genuinely open. Google could never be open about its search algorithms, not simply because Bing would instantly be as good as Google but also because people would use that information to game the system.

And that’s the issue: Having invoked the magic “open” word, you’re a hostage to fortune. Any time that the rational decision is “don’t be open” (as it is, arguably, with Honeycomb’s source) sneering naysayers like me will be on your case, whacking you over the head.

Enhanced by Zemanta

Why free software will remain a niche, in a nutshell

Steven J Vaughan-Nichols on The new Debian Linux: Irrelevant? | ZDNet:

“For example, the default Debian distributions won’t include any proprietary firmware binary files… If, as is likely if you’re using a laptop or a PC with high-end graphics and you find you’re running into hardware problems, the Debian installation program should alert you the problem. That’s fine as far as it goes, but the installation routine won’t automatically download the missing firmware from the Web. Instead, you’ll need to pause the installation while you fetch the missing in action firmware from either the Debian non-free firmware ftp site or the vendor’s site.

The theory is that by doing this outraged users will demand that hardware vendors will open-source their device drivers, or, at the least, let Linux developers write open-source drivers for proprietary hardware. In practice, it doesn’t work that way.”

While the commitment to free software inconveniences users who want to mix-and-match how much they use it, it will remain a niche choice.

How efficient is open source development?

Mark Shuttleworth, who probably knows more about open source development than most, on the open source development process:

“There is a myth that being open is necessarily more efficient and cheaper, but there are no hordes of people showing up to do the hard stuff,” Shuttleworth says. “Occasionally wonderful, magical things happen — really incredible things do happen, like people show up unexpectedly with brilliant ideas — but it’s still hard and expensive and you still have to be willing to do all the hard and expensive things and do it in an open fashion. And you’re still likely to be accused of being open only when it’s convenient.”

Eric S Raymond’s formulation of Linus’s law proves not to be true. Not so much a law, more a statement of desire, then.

[EDIT: Oh, I forgot to mention – this is the first post composed and sent from my iPad :) ]

Reblog this post [with Zemanta]

Create things people want to own

Sometimes, I start writing a blog post only to find that someone else has said what I was about to say, only with more brevity and style.

In this case, it’s Joe Clark, talking about the open source community’s “sky is falling in” act about the iPad:

“This was the weekend those of us with high standards lost their remaining residue of patience for ideologues who hyperbolize about open systems without actually creating something people want to use.”

Sometime, I will actually get around the writing my post, though. I promise.

Reblog this post [with Zemanta]

Daring Fireball’s wishful thinking

I totally understand where John Gruber is coming from with his post on “The OS Opportunity“. The problem is that there’s a whole lot of wishful thinking in there.

First of all, go read John’s post. Rather than try and summarise here and potentially mischaracterise what he’s saying, you just go read it.

Back? Good. Then we’ll begin. John’s first point is that what kept people on DOS was simply file compatibility:

“In those days, before DOS ran most competing platforms out of the market, interoperability and data interchange were at best difficult, and often impossible. Data was stored in incompatible file formats written to incompatible floppy disks1 by incompatible apps compiled for incompatible CPU architectures. Even later in the ’80s, when networking became common (at least in businesses) the network protocols were proprietary.

That was the world where DOS won out. Get everyone on DOS and you could all open each other’s WordPerfect and 1-2-3 files, if only by sharing them on floppy disks. So DOS gained users, and because it gained users it got developers, and because it gained developers it got more users.”

While this is partially true, it ignores two other factors which always mitigate against switching platform – and which continue to do so today.

The first is familiarity. Familiarity, to geeks like me and John, is something you often avoid like the plague. Geeks like us like tinkering with new stuff, learning how to do new things with new tools. We switch because it’s fun (today’s example of this from John: Switching to Camino. Only geeks like us look at switching browsers as the kind of thing you can do on a whim. Why else does the blatantly inferior IE retain so much market share?)

But for someone with years of experience of DOS (or Windows), running WordPerfect and Lotus, switching to an alternate operating system and set of applications was always a big deal. The path of least resistance was always to stick with the platform you’re on, because learning new stuff got in the way. GUIs mitigated this a bit – but didn’t change the situation with applications. For someone who’s been using Excel professionally for 10 years, switching to Numbers is a big, big deal – and that’s despite Numbers being pretty easy to get your head around.

This is even more apparent in the business world, where switching means training hundreds of users in how to use the new tools. There’s a very good reason why corporates tend to be a couple of versions behind the latest, even for products where there’s a clear, delineated upgrade path and a level of familiarity.

The second reason is the oldest one in the book: money. If you’re a seasoned Windows user, switching from Windows to Mac doesn’t just cost you the time to learn new applications (even when there’s a Mac version of a Windows app, they’re usually different enough to cause angst). You have to actually buy the applications, because few (if any) companies give freebies to switchers.

Of course, this second issue isn’t an issue if you’re switching from closed source to open source. And some of it is also negated by being able to use freebie tools on the web. But the more complex your needs, the less likely it is that either can fill them. And the quality of both free online tools and open source stuff is (to be kind) variable, particularly when it comes to the kind of simplicity of interface design that someone switching OS’s is going to appreciate. I know – I’ve done it.

“A similar feedback loop is going on with the iPhone today, but it’s far less sticky. The DOS/Windows monopoly grew impregnable because it was a platform where the only way to play along was to join it.”

John’s right that this feedback loop is going on with the iPhone, and that it’s less sticky, but there’s two reasons for that. First, the smartphone software market is nascent: it’s in the equivalent of the era of (as John puts it) “the Apple II, the IBM PC and DOS, Commodore, Atari, Acorn. The TI-99/4A.”. People forget that DOS wasn’t the only game in town – only the weight of IBM’s brand and the anti-trust rules which allowed Compaq and a slew of others to clone the IBM PC really made it the overall winner. Even the iPhone, which is massive in terms of mindshare, only has 17% of the smartphone market. That’s about as much as the Apple II had at its high-point. The smartphone market is still massively fragmented – and it’s a very open question whether that will continue.

John’s bet, I think, is that it will continue to be fragmented – although I don’t think he overtly states this, so please forgive me if I’m reading something in that’s not really there.

I think that assuming this is true says a lot about what you believe is the future of mobile software. If you think that smartphone software is fundamentally one-trick apps, throwaways, stuff which is easy to develop and easy to dump, then jumping from one smartphone to another is always going to be easy.

But if you think that developers are going to create more and more complex apps, and that these are what consumers will increasing demand and use, then switching becomes more of an issue. The fact that Omnifocus is only on iPhone will almost certainly mean that my next phone will also be an iPhone, despite my constant pain at the fact that the iPhone doesn’t multi-task. If there was no Omnifocus, I would switch. And I suspect that I’m going to be increasingly not alone – with 100,000 apps, the potential for the “just one app that I need” being on iPhone grows.

“If Palm can create WebOS for pocket-sized computers — replete with an email client, calendaring app, web browser, and SDK — why couldn’t these companies make something equivalent for full-size computers? The hard part of what Palm is doing with WebOS is getting acceptable performance out of a cell phone processor.”

Because no one would buy it. It’s not like people haven’t tried. There’s a very good reason why people have chosen Windows netbooks over Linux ones, even when Linux has been cheaper – they want to run the apps they are familiar with. And they don’t generally just want web apps – they want native ones. Rich beats thin, every time.

“These PC makers are lacking in neither financial resources nor opportunity. What they’re lacking is ambition, gumption, and passion for great software and new frontiers. They’re busy dying.”

And this is where John’s wishful thinking really comes to the fore. Who, exactly, is dying? HP, which made $2.2 billion profit in its last quarter? Dell, which made $472 million profit? While those aren’t as good as Apple’s numbers (because SteveJ has played a very smart game), neither looks like a company that’s “busy dying” to me.

Reblog this post [with Zemanta]

Windows 7, Mac OS X and Ubuntu: A Tale of Three Operating Systems

Joe Wilcox picks up on a comment that I made on his post about Windows 7 and its relationship to the Mac:

“As you know, Joe, I’m a Mac to Linux switcher (with over 20 years Mac use under my belt). But I’m also a tinkerer who’s curious about OS’s, so I’ve been running Windows 7 as my main system for a month or two. Count me amongst the impressed. Microsoft has actually applied some real serious effort to the user interface design, taken some of Apple’s ideas, and made them better. That it’s much, much faster than Vista is a bonus.

Mac fans should take a serious look at 7—not because it will persuade them to switch, but because it’s the first serious competition from Microsoft in quite some time.”

Joe’s timing is impeccable, as in a couple of weeks I’ll be switching my main computer back to Ubuntu from Windows 7. But the reason isn’t exasperation with Windows 7, and it’s not one that should give Mac fans hoping that the new Microsoft OS will be a failure any kind of comfort.

Continue reading

How much of a success is open source?

How much of a success is open source? In his musings on open source, and how ideas cross the chasm, Alan Patrick ponders the origin story of open source, and how it relates to a particular brand of utopianism.

“The problem of course, is that many of these Utopians are the dreamers and idealists who got in early and inspired so many others to join the movement in the first place. Without these enthusiastic early adopters, these ideas would never get off the ground to be in a position where the leaders do have to grasp the nettles.”

Part of the problem, too, is that too many promises were made by open source evangelists who understood neither project management nor people management. Anyone who’s even passingly familiar with project management knows that piling more “eyeballs” on a problem doesn’t make it shallow: what you need are the right eyeballs, in the right context, at the right time. This becomes more and more true as projects become deeply complex: someone picking up the code of, say, MySQL today will have quite a long learning curve before they can meaningfully contribute to the project.

Continue reading