Jim Minatel's Wrox Book Editor Blog ASP.NET, XML, CSS, Ajax, PHP

Google Book Search is a Safari Killer

There's been a lot of hand-wringing about the the effect of Google Book Search but I think the real loser in the recent publisher's settlement with Google stands to be Safari. If you think about the way you read print books, or even ebooks in PDF or Kindle, you're probably doing a lot of cover to cover or at least full chapter reading. But Safari (and Books24x7 for that matter) seem much more usable as "search" tools to find a specific answer to a specific need. In using either of these, I'm much more likely to need to read just a few pages, up to maybe a chapter at most.

Enter Google Book Search. Going forward, you're going to see a lot more books in this designated "limited preview." Rather than thinking of this as a limit, I think readers will quickly discover that here's a new way to get a few pages of answers from many books. From as many or books more than Safari for example. Google Book Search does let the publisher of currently copyrighted books set how much of their book can be "previewed" by a reader in a set period. Most publishers will set that at around 10-20%. For a typical 500 page computer book, that's going to give readers 50-100 pages to read free.

And I think, that has to be more of the typical book than the typical Safari reader reads from any one book.

So what about book choice and selection? There it already seems Google is using their muscle to outpace Safari too. For computer books, the books published in the last few years are what most people read most.So I did quick comparisons of a few key topics, how many books published from 2005-2009 are available in Google Book Search vs Safari:

• JavaScript: Safari 37 books with JavaScript in the title vs 36 in Google (english language). Google includes some of the bestsellers, such as both Goodman books (Wiley Bible and O'Reilly Cookbook), both Wrox Professional (by Nicholas C. Zakas) and Beginning (Paul Wilton), the Crockford "Good Parts" and so on. You'll get an Addison Wesley Dojo book and the recent Wrox jQuery (Richard York). In short, if you have a question or problem about JavaScript that a book might hold the answer to, you've got just as good a chance finding the book with the answer in Google than safari.
• ASP.NET 3.5: Safari 14 titles vs 20 in Google (english). Here Google has all the expected bestsellers (our Wrox Professional by Evjen, Hanselman, and Rader; Wrox Beginning by Imar Spaanjaars; Sams Unleashed, and so on) and even the most recent Professional ASP.NET MVC 1.0 by the new gang of fourheads (Conery, Hanselman, Haack, and Guthrie). There's nothing on ASP.NET MVC in Safari.

I'm sure there are books and topics that Safari (and Books24x7) users will find they need, or need more of, that they can't get or get enough of from Google, but will that be enough to continue paying $23-$43 a month for Safari? Or will the additional note taking features in Safari be enough to justify subscribing?

If you read the O'Reilly Radar blog and in particular, posts in the last few years about the "state of the computer book market," it's no surprise that the market for selling computer books is tough lately. One of the few bright spots in the industry has been Safari, which isn't accounted for in the print book sales "bookscan" data in the market reports. And if Google Book Search makes a big dent in Safari and other similar subscription services, that is bad for readers, authors, and publishers. For publishers who are counting on Safari or similar services to subsidizes investments in content (that is paying the authors and editors who create the books), it's not a good sign. And it probably means you'll see less investment in innovative services like Safari (which O'reilly, Pearson, Microsoft, have been building for more than 10 years).

Kindle Source Code Release: What Does It Mean?

This is an interesting surprise to me. Amazon released their source code for all the Kindle devices. What does it mean?

• Will we see competing hardware devices running clones or forks of the Kindle source?
• If we do see 3rd party devices, will their users be able to buy from the Amazon Kindle store?
• Will we see Kindle software solutions for existing hardware platforms? Amazon's already released Kindle for iPhone (which I use). Will we see other vendors use it too?
• Will this allow other retailers to sell Kindle ebooks from their site directly to other devices?
• Could new Kindle clones or stores mean DRM-free Kindle books?

This is surprising to me when I try to look at it from Amazon's business perspective. They controlled the device and the channel in the existing model, I'm not sure what they think they had to gain. I wonder if this is in some way tied to the Kindle DX textbook trials announced recently, and maybe they felt that they needed to open the platform to avoid the appearances of the beginning of an ebook monopoly?

From a customer perspective, it's a huge win if this drives down the cost of Kindle hardware. And maybe Amazon's win is making the hardware ubiquitous at someone else's expense to drive more Kindle book sales.

As always, these musing are purely my own. Even though I work in a company that sells a lot of books through Amazon, I don't have any direct knowledge of their business decisions.

Updated 6/17/09: Thanks Eric for pointing out (see comment) that the original Kindle and Kindle 2 source have been available a while and that all that's new here is the DX code. I wonder then if anyone knows of any 3rd party apps, devices, or sites making use of this source yet?

Twitter Needs to Evolve

Imagine this: you go to send an email to a new contact you met at a conference and are told by your email server that you have to many email contacts and can't send this message until you delete some contacts. Or imagine that your all of your company's email is dependent on one external service provider, the universal email service provider for the world.
That's unfortunately where we are with Twitter and why Twitter needs to evolve, again. I say again because if you look at the history of twitter, it's already evolved and evolving, but it will need at least one more big step on the evolution chain. That Twitter history would be:

1. Twitter was launched primarily an SMS service for mobile phones. Although there was a web interface, the primary expectation was you'd get updates from a few dozen friends as text messages on your phone. I remember using Twitter "back then" around the Microsoft Mix conference in 2007 to stalk, er follow, authors and prospects via their Twitters to instantly know who was in the speaker lounge or playing video games in the hall. (This was in my old @jimminatel days before the @wrox account passed to me.)
2. Twitter's first big evolutionary step was to de-emphasize the sms/txt dominance and grow toward the web being an equal interface.
3. The introduction and growth of the Twitter API and various clients and applications based on it marks the current state of Twitter evolution. While many casual Twitter users still use the web interface, the leading edge are definitely migrating to client (phone and desktop) apps.

The Future of Twitter Is...
Twitter's next stage of evolution needs to look at the API usage as the future and build a service around that. That evolution probably shouldn't mean growing city-size data centers to house what will be en ever growing need for server space. Instead, Twitter needs to de-centralize. I see one of two possible paths here:

1. The first option for Twitter is to become something like e-mail, where there are common standards for internet exchange of email. Any entity that wants Twitter access would set up their own Twitter servers according to their needs or purchase access from others. Twitter's own corporate business models in this could be selling the first software to run these, or hosting a free service supported by ads, or selling premium access as has been speculated lately. This option would nicely allow for sites/servers running Twitter to take on domain names like the rest of the internet and for corporations to have corporate twitter addresses.
2. Another option might be more like many of the current peer-to-peer file sharing networks. It's hard for me to see how Twitter as a company can directly benefit from hosting and routing all Twitter traffic on servers they pay for. I can see a community of distributed p2p-Twitter services helping distribute this load.

Either of these schemes involves some fundamental rework of how Twitter works, and neither is a Trivial task. Neither completely solves the issue of how Twitter successfully makes money from their creation either.

But if Twitter becomes as popular as email, instant messaging, or text messaging (and why shouldn't it be that popular?) the current architecture seems unsustainable.

Why the WSJ Gets it Wrong on Ebooks

There's much excitement and retweeting today on the WSJ article How the E-Book Will Change the Way We Read and Write. It's all based on a faulty premise that the majority of the population would with instant access to 10's of millions of books do more reading like the author Mr. Johnson does.

Sorry to break your bubble Mr Johnson but here's what the internet age has shown us sells when instantly available:

• the 1 billionth itune sale alongside uncountable billions of free music downloads

• Video games

• Fart noises - want your iphone app to sell 100x more? Just add fart

• Movies, and their cousin TV shows

• And of course, porn.

The world is willing to pay for mostly passive entertainment, with the minor exception of video games and their false sense of active participation.

What we book people are often lured into forgetting is that we're a little different. We'll sacrifice the instant entertainment of a 2 hour movie for 6 hours reading the thing in print. We're living anachronisms, along with our children whom we shower with books in hopes they'll carry on our holy mission of reading.

Why Doesn't Amazon Want Paid Google Search Customers

I'm completely stumped by this. Earlier this week, Amazon announced to their North American Amazon Associates that they would no longer pay referral fees to Associates for users sent to Amazon through paid searches on Google, Yahoo, MSN and other search engines.

I'm trying to understand the business logic here. If an Amazon Associate spends money advertising on Google and other search engines, and that paid search leads to a customer buying from Amazon, why does Amazon value that transaction less to the point of not calculating it for Associate referral fees?

Can anyone explain this?

Learning from Futurama

"Good news everyone!" Fans of Futurama don't need any explanation.

As I wait for my next 2 Futurama movies to arrive on DVD, I've been pondering the future of free TV, and with it, the future of free content vs paid content everywhere.

Futurama was of course canceled by Fox after 4 seasons. Although Fox bounced it around the schedule and often preempted and delayed it for football, we watched it religiously. And it had a weekly audience of 6-10 million viewers. That wasn't enough ad revenue apparently to pay the bills and keep it in production.

But, a few years later after starting syndication, Futurama did something brilliant: "Good news everyone." Futurama began a series of direct-to-DVD videos. As of today, there are 4 DVDs in the series, and at roughly $20 bucks a pop, I've bought all of them. That's $80 off me in moves, and I'm sure me watching the original 88 episodes on TV didn't bring them a buck a pop per episode in ads. So they're making a lot more from selling me DVDs than from me watching free TV.

Of course, not all casual Futurama viewers would pay for the movies so I've no idea how the net on each movie compares to the net on 4 episodes. (That's the right comparisons because the movies are each roughly equivalent to 4 episodes and in fact, are being split into "episodes" for the cartoon network.)

I've also been thinking about this this week because Fox (the villain again) is pulling the plug on Sarah Connor Terminator Chronicles this week which after 2 seasons is suffering from chronically lower viewers. Fan sites are adamant that actual viewership is higher than the official ratings when DVRs and online viewers are counted. But since no one with a DVR watches commercials, those arguments are meaningless and Sarah is "hasta la vista baby."

So what's the future of free TV?

• Live events especially contest oriented shows will be safe longer because they defy DVR viewing

• Shows that appeal to an audience that isn't DVR-savvy show some promise. Anyone for more Nash Bridges?

But if your audience is technically literate and plugged in, that very technology may be the undoing of the good content it thrives on.

Fortunately, computer book authors and online article authors don't have to deal with DVRs. And although online ad blockers exist, it seems the biggest problem may be the ambivalence of web surfers to online ads.

I haven't decided what I think all this means for the state of free versus paid content in other forms, like computer book content. What I have seen in general suggests that the same author writing the same content can generate more paid revenue than that same content generates in ad revenue.

What do you think?

ASP.NET MVC Open Source is Good

Earlier this week Scott Guthrie announced that ASP.NET MVC would be released Open Source, under the Microsoft Public License (MS-PL). While the community response was generally good, I was surprised by some of the people I saw complaining that this isn't good enough. I've had a few days to watch the responses and think about this, and here's why I think this is good and Microsoft has done the right thing and the complainers should find another gift horse to look in the mouth.

1. First, most of the responses that I have read are positive. Even on the usually unfriendly ground at Slashdot many posters have defended and supported the decision and the MS-PL. The mood's been mostly positive about this at dreamincode. The in the twitter community I haven't kept a formal count but I've seen more "this is good" posts than not.

2. But the first complaint that I see is: Why use the MS-PL and not an "accepted" open source license? Well first, MS-PL is an OSI approved license. So if not MS-PL which open source license would have been acceptable? BSD? Apache? GNU? (Which GNU)? MIT? Why is it OK that PHP and Python can have their own open source licenses but Microsoft can't? Why can Mozilla have their own open source license but Microsoft can't? Oh, yes, this is Microsoft so we begin with an assumption of guilt.

3. Another complaint is that while ASP.NET MVC's source has been released as open source, it isn't actually being developed like Open Source. If it's good enough for "fill in the name of your favorite open source project, why isn't it good enough for Microsoft?" My thought on this is, why does every open source project have to follow the same set of rules? They don't. If you like the way Microsoft is working and releasing ASP.NET MVC open source, great use it. But if you don't like it, don't say it's no good and it isn't really open source just because they haven't chosen to let you commit your changes to the code to their code base.

4. "Microsoft's running scared, hemorrhaging developers to RoR and this is their attempt to muscle in." Hmm. Even if the hemorrhaging users developers comments were true, wouldn't it be refreshing to see a major company that actually adapted and changed their business practices when they saw their audience needs change, when a "competitor" did do something better? Maybe if that were the case we'd have seen some automakers changing a lot faster 20 years ago and a lot less in need of bailouts today. But the truth is, ASP.NET itself has a huge developer base. So does PHP. RoR has a spot. As do many other technology stacks. None of those 3 are going away any time soon. But I expect all 3 of them will evolve. Even David Heinemeier Hansson who's built a reputation for building Rails for his company's needs, showed an evolution in the decision to merge Merb and Rails. Or put another way: if the decision is good for developers, as the ASP.NET MVC open source decision and Rails-Merb mergers both are, why can't we all just be happy and accept them?

So, congrats to ScottGu and the ASP.NET team for the ability to make this decision. Congrats for Microsoft for allowing ScottGu to make the right decision. Not all big companies are this accepting of new ideas.