the onda

One of the great things to come out of Apple's snub of Flash during the iPad launch has been the discussion around whether Flash is really needed for a first-class web browsing experience. In reading the various opinions, I came across a brilliant Safari plug-in, ClickToFlash, which shims the Flash player so that you have to explicitly approve any Flash widgets that are loaded on pages. The plug-in provides a whitelist so that certain domains will always load Flash (say Pandora and Hulu) widgets. Apparently there are also versions of this same general idea for Firefox and Chrome.

I've been using ClickToFlash for 4 days and there is no going back for me. Apple's claim that the Mac crashes most because of Flash seems well supported by the fact that I can now have 20-25 tabs open regularly without Safari hogging all of my machine's resources. It's awesome.

What is most nefarious about Flash is not all of the explicit video embeds that people use, but the fact that it has snuck into display advertising. If you install any of these plug-ins, you'll quickly realize how many of the standard IAB ad sizes are now actually being powered by Flash widgets— which do nothing but pulse in annoying ways— and perhaps more significantly— suck down system resources.

Just say no to Flash.

I wanted to wait a week to put down my preliminary thoughts on the iPad both because of the crazy outpouring of people venting some very real emotion on all sides of the debate, and because I've learned that when it comes to Apple and Steve Jobs, it often pays to let the reality distortion field fade a bit.

I am not a fan of netbooks— as I've written before, they are nothing but cheap laptops. Thus I was sort of excited to see Apple try to reinvent the category with the iPad. After a week of reading reviews though, I'm sad to say that this seems like a fairly incremental product— taking the best parts of the closed ecosystem of the iPhone/iPod touch and hitting the 4x magnify button on the form factor. It'll be a fun and possibly lucrative product— and in the end I could see Apple selling 1/3 to 1/2 as many of these as they have iPod touches (which is about 40MM to date). But here is where they fell short:

1. New formats: the Kindle sucks not because of its screen or hospital ID but because despite the connection to the Amazon web service and the persistent data connection, it fails to be as good as a paper book and doesn't reinvent any of the parts of the book that could use some social/cloud juice. From what I could see, Apple is doing nothing better here (at least with version 1). They haven't defined a new format that would allow publishers to take advantage of video and audio and they didn't do anything to make the experience of finding, reading, and sharing books any more social. Could be that the publishers weren't hip to it, but if anyone could crack that particular cartel, I thought Apple could.

2. Ergonomics: Ned wrote this one up, and frankly, I can't believe the Reality Distortion Field has kept other people from noticing that the idea of a non-tactile keyboard without a place to rest one's palms renders this a piss poor input device. This is ironic as Apple was the first company to do the palm rest design for laptops with the Powerbook 100/140/170 series that was even shown during the keynote.

3. Closed ecosystem: I'm less worried about this one than the people who feel that Apple is starting us on the road to being like the bio-sacks that pass as humans in the movie Walle, because it does seem as though the combination of HTML5 and the A4 CPU will make webapps really sing on this machine, but it would have been nice to have given people some sort of mechanism for side-loading applications (even as "untrusted" like Android and S60 do). Again, this is ironic coming from the company that had its first hit product (the Apple ][) succeed because some crazy hackers in Boston wrote a killer app and distributed it with little involvement from Apple. That said, I do think if the iPad becomes ubiquitous enough, Apple will have to open up (or be subject to aggressive jailbreaking as the FUD around "messing with the carrier settings") doesn't exist for this type of device.

Finally, perhaps the biggest disappointment for me is that because of 1-3 the iPad seems to be primarily a content consumption device, a sort of modern-day version of the portable TVs that Sony and others started selling in the 1980s. That may be what the market wants, but I find it hard to muster the same level of excitement as I would have had for a more general computing type of device. In the case of the latter, creativity really is the upper limit on what can be done. With a closed, content consuming device, we'll end up where we are with the iPhone: with small widgets that all basically do the similar types of information retrieval and display. If this is the new world of personal computing, it looks a little boring.

The humble and most underappreciated client of the RSS revolution, Google Reader, has just launched an interesting feature: the ability for a user to subscribe to a page that doesn't natively emit a syndication feed and still receive changes as a stream of events.

As RWW reports, this is not the first time a product has attempted to do this (and more significantly, geeks have been scraping webpages since the dawn of Perl), but the fact that it is Google, doing it at Google scale, is particularly relevant.

For a while now, we've had available both the page-oriented web that is best suited for humans to consume, and the stream-oriented publishing formats that blogs and RSS pioneered and that Facebook and Twitter popularized. In the middle, a few services have willing to take the step of translating the torrent of the stream into something page-oriented that people can consume in a page format (look at Techmeme or Tweetmeme for two great examples), but the promise of the programmable web is to lower the required investment in making a million such efforts not only possible, but easy.

And there lies the rub: Google providing any webpage's diffs as a stream is bound to lower the barriers to anyone looking to build on top of streams to create their own page aggregators. This was the promise of Yahoo Pipes (or Google Mash Editor), two products that couldn't overcome the complexity of incorporating content that was not readily digestible in one of a few syndication formats.

It's easy to imagine the possibilities, especially because we've got a good set of aggregators that scrape e-commerce, travel, and finance sites already. However I suspect the real promise of this particular Lego brick will come in scraping public data websites that have previously been ignored, either at the local level or around particular topics.

And I for one would much rather see Google spending engineering cycles in useful extensions to the web like this one than on bashing it out with Apple over the next dominant consumer computing device.

Chris Anderson has a piece in the latest issue of Wired, "Atoms Are the New Bits" (not yet online) which chronicles the emergence of DIY hardware development. Covering the basics of prototyping tools like 3D printers and CNC machines, he goes on to make a connection to the newly emerging short-run Chinese manufacturing supply chain to explain how a new industrial revolution might take place, even going so far as to quote from my favorite fiction book of 2009, Cory Doctorow's Makers:

The days of companies with names like "General Electric" and "General Mills" and "General Motors" are over. The money on the table is like krill: a billion little entrepreneurial opportunities that can be discovered and exploited by smart, creative people.

As with his article on the Long Tail from five years ago, I think Anderson is on to something very big here even if he doesn't get all the causation correct. Short run manufacturing could be the basis for a whole new industrial base here, but not necessarily because of the democratization of prototyping tools (which as the story of Makerbot Industries shows is indeed happening along the same story arc as the PC revolution), but because of the powerful combination of the Internet as a collaboration channel (for ideas) and the Internet as a demand aggregation and distribution channel (for niche interests, passions, and ultimately sales).

This is the story of Threadless, a company which exists today because it is perfectly evolved to take advantage of these two forces in the apparel space. But increasingly it may also become the story of much more complex products. The Wired piece covers a local Boston startup called Local Motors which is trying the approach with cars— a complex multi-component electromechanical product that has all sorts of safety and regulatory challenges. If they can pull it off there is no reason not to wonder about other similarly complex products.

For instance, just the other day I was commenting to someone that I'd love to have a smartphone that had an internal 3000 mAh (they mostly have 1/2 of that) and a cheap low-power 7-segment display for most of the notifications that it would normally power up the battery for— "the longest lasting smartphone in the world." One can argue whether such a monster might end up being the Spruce Goose of the category, but with Android we've got the software to be able to do just that, and looking through any of the iSuppli teardowns, you quickly realize that most of the internal components of these devices are increasingly as "standard" as what goes into your typical PC.

How far are we from the day when someone can start the Threadless of smartphones? Then we could go from custom cases and wallpapers to devices truly tailored to our specific needs.

There are many wrecked ships that have fallen victim to the siren song of mass customization, but given enough relevant customization in a product category that has enough demand and you may just have something here.

Amazon opening up the Kindle with an SDK strikes me as an incredibly stupid move, or at the very least, one which is more based on the current wave of hype around SDKs or fear around the impending Apple tablet. Having been part of many "app platform" discussions over the last couple of years, I can't help but feel that most people need to stop reading the iPhone informercials that pass as "analysis" in the tech and business media and remember that most devices really are just appliances and don't need the overhead and potential instability of letting 3rd parties get up close and personal. And that most app efforts require developers, a generally non-bozo crowd that won't sink precious time into things that smell bad just because they looked good on a Powerpoint slide outlining the "stickiness strategy."

For instance, would you want your toilet to have an SDK so that third parties could build apps that listened for the onflush event and did fun things therein?

What really matters is that these connected appliances be extensible in ways that make sense, not that each one brings some new proprietary app environment that developers will have to learn. In the case of the Kindle, Amazon achieved this brilliantly with the "email to Kindle" feature that allowed anyone to send in a list of pre-approved document types for display on the device. I would even argue that for most smartphone platforms, the extensibility should come form the capable HTML5 mobile web browsers that allow for very specific app-like experiences without going down the SDK rathole.

It is app store mania continuing its mad rush, and I for one don't get why we are so ready to jump back into the days of developing with different libraries, toolkits, and operating systems like we did in the pre-web days.