Simon’s Blog

MacDailyNews has posted a rumor about a forthcoming Mac Tablet computer which they call a MacBook Touch. People can call it a rumor if they like, but I think it’s more like a dead-cert. Why? Because tablets are now a compelling value proposition for people - today, the technology works really well; and is available to build at reasonable prices.

How do I know that tablet-like technology can work so well today? It’s because Microsoft has already made a reasonable start. Not many people know this, but “tablet features” are actually built into each and every copy of Microsoft’s Vista. If you spend a few tens of dollars on a good quality graphics tablet (yes, that’s all they cost these days), you can use a pen to input data - writing or drawing - into your Vista notebook or desktop. The hand-writing recognition is a revelation. If the computer can understand my scrawl with near 100% accuracy (which it can), then it can understand anyone’s writing. With zero training. As for drawing, any graphic designer will tell you that a pressure-sensitive graphics tablet is a great way to use a computer for creating fabulous graphic assets. Of course, Microsoft has been trying to push the Tablet PC form factor for a while, but the products haven’t really caught on so far.

So, here’s what I expect Apple might do, if the leadership is feeling really bold: replace the entry-level MacBook product line, with an all-new MacBook Touch product line. I’d guess at the following key features:

• There will be a stylus - inputting text via hand-writing on a large surface works really well, and feels completely natural with fast, modern hardware
• It will be a multi-touch screen, that will work with fingers, as well as the stylus
• It will be a pressure-sensitive screen, to make drawing work well (e.g. thicker lines, when the pen is pressed harder)
• There will be a keyboard, and some kind of funky hinged, rotating screen - like Microsoft Tablet PCs, but engineered to higher standards, so the screen hinge feels really solid
• The software will be better than Vista tablet (which is already quite good), with superior integration of touch (fingers, and stylus) into the OS.

For sure, tablet style notebooks are now ready for prime time. If Apple does something similar to the above, it will be Apple and not Microsoft that takes tablet personal computers into the volume mainstream…

It looks as if Sun will be releasing the first technology preview of the all-new JavaFX SDK in about two weeks - that is, at the end of the month.    That includes: a new language (JavaFX Script); a new compiler; a new underlying Java hardware-accelerated 2-D and 3-D graphics stack;  (hopefully) new audio and video support;  a new Java plug-in and other features of Java6 Update 10.

That’s a huge amount of work - and it looks like the team still has more to do before the end of the month. Anyone that’s had to ship complex software to hard deadlines before  will be wishing them a lot of luck. It will mean working really long hours, and there will be huge amounts of stress.   I’m sure they can do it though - in situations like this, previously untapped reserves of energy and brain capacity seem to kick in, and the work rate gets crazy-fast.

After they’ve shipped this first technology preview, and the dust settles, I really hope the JavaFX team will take the following points on board.  I think they are critical to the success of the JavFX project:

1) JavaFX demos need to be applets, not Web Start applications

JavaFX is all about the consumer experience.  Web Start is horrible from a consumer point of view - it involves clicking on buttons in ugly dialogue boxes with horrible warning triangles and silly messages about trusting or not trusting the publisher.  So - no Web Start please! If JavaFX is to be a serious consumer RIA player, then JavaFX  RIAs need to “just start” seamlessly, and rapidly, as applets embedded in web pages.   Just like Flash RIAs do. Just like AJAX RIAs do.

Making all demos applets will force the JavaFX team to eat their own dog food.  That is, use the technology as developers will use it to build real RIAs.  If the technology isn’t good enough to build pretty much all demos as applets, then what hope is there for real-world RIAs?

2) The fact that the content in a web page is a Java/JavaFX applet needs to be invisible the the user

When Flash or AJAX RIAs start up, the end user sees nothing to interrupt their browsing experience.   It’s just like visiting any web page.  Currently, when Java applets start up, some really horrible things can happen.   You sometimes get a BIG ORANGE advert for Java and Java.com in the web page.  You always get a Java icon showing up in the system tray, when the Java VM starts up.   I’ve seen some Sun people arguing that neither of these things matter.   Well - those people are wrong.   These are both significant net negatives for the end user.     From the end user perspective, the RIA is just another part of the web page;  it is not special enough to justify anything different happening when the Java/JavaFX stuff starts up.

To be honest, the fact that some Sun people don’t understand this, worries me; and makes me concerned that they don’t actually understand what it’s going to take to build a compelling consumer-facing technology.   The point is: this is really basic 101-level stuff - no-one should have any trouble understanding it.  A helpful rule of thumb for anyone that doesn’t understand the fundamentals is: if the behaviour I build in is different from the way Flash behaves, I’m  probably doing something wrong.   For example, when Flash content starts in browsers - do you see big Adobe/Flash adverts?  NO!  When Flash content starts in browsers, do you see an Adobe/Flash icon in the System Tray? NO!

3) Get the JavaFX Designer Tool Shipped ASAP, For Free

In 2008, sensible people don’t hand-code the bulk of their user interfaces.  They use visual tools to do much of the heavy-lifting i.e. auto-generate code.  For example, with Java Swing, sensible developers use Matisse in Netbeans.   So - going back to hand-coding UIs with JavaFX is going to be a retrograde step that will limit adoption. Sun is working on a visual design tool for JavaFX.  It’s aimed at graphic designers, but developers will also  want to use it for majors part of their work. The point is: it will be the only sensible way to build UIs using JavaFX.

So, the truth is - without a great visual design tool, it will almost certainly be more productive for Java developers to continue to build Swing UIs by using Netbeans/Matisse, than it will be to build JavaFX UIs.  Clearly, the JavaFX Design Tool isn’t going to be finished before it’s finished.  However, the sooner it’s ready, the sooner Java developers will start adopting JavaFX.  And, of course, the design tool needs to be free, just like Netbeans.

I think that’s it.   It’s a really exciting time in the world of client Java.  I’m looking forward to seeing some amazing new demos when the JavaFX SDK technology preview comes out!

There’s no doubt about it - with iPhone V1.0, and now iPhone 3G, Apple caught Nokia asleep on the job.  iPhone changes everything, and Nokia’s lack of credible response suggests that their senior management is: a) in denial about this; and b) doesn’t know how best to respond.  Apple is, today, a small player in the mobile world.  However, having fixed the critical business and technical problems with iPhone V1.0, Apple is now on a steep trajectory that will see them become a major player in the mobile world (update - it took 74 days to sell one million iPhone V1.0s; but just 3 days to sell one million iPhone 3Gs).  How should Nokia respond?  Herewith, what I think the top three priorities for the company should be.

1) Build A “Better” iPhone

Is iPhone the perfect mobile device?  No, it’s not.  That means, there’s an opportunity to build a phone that’s better than iPhone.  What’s now the No. 1 problem with iPhone?  The touch screen keypad doesn’t work well enough.   Nokia should build a phone that is exactly like the iPhone (functionality, user interface, sensors, quality of materials, dimensions, screen resolution and size etc.)… you know, match the iPhone on every level… except it should also have a great quality slide-out QWERTY keyboard.   Oh, and while they’re at it, they can double the storage, put a faster CPU in it, and include a better camera.  Is that a technical challenge?  Yes - but what do they pay their R&D people for?   It’s not a difficult brief to understand - build an iPhone with a great slide-out keyboard.

2) Take Control Of Phone Software/Firmware Updates

Nokia needs to recognise that Apple’s control of iPhone software/firmware is a winning strategy.  Mobile phone operators have no clue how to provide a good service to their customers.  They never will; and the truth is they have no desire to.

Once Nokia has done this, they then need to focus on improving the device over time. Nokia currently ships phones with crappy, unreliable, buggy OSes.  Nokia got away with shipping rubbish when all the competition did the same.  This is not good enough now that Apple is bring quality software to mobile phones.    Time to start dedicating the proper resources to shipping super-high-quality software.

3) Take Control Of Deployment Of Third-Party Applications

Nokia needs to recognise that Apple’s strategy of controlling deployment of third-party applications to phones is a winning strategy.  The mobile operators are clueless about this (see point 2 above).   Nokia needs an App Store, just like Apple’s.  No creativity required.  Just copy Apple.

Bottom Line

Nokia needs to play catch up with Apple, and then overtake it.  It’s not difficult to do: copy Apple’s winning strategy, and ship hardware that is superior by addressing iPhone’s key weakness.   What are they waiting for?

So, will Nokia implement the above changes?  I doubt it.  My bet is that they will continue to cede control of their devices to useless mobile phone operators that don’t care about customers.  And their response to iPhone, when it arrives, will be a poor, slow, cheap plasticky imitation with crappy buggy software that delivers an infinitely worse user experience compared to iPhone.  In other words, I suspect Nokia’s management is still in denial about iPhone; and they still don’t know how to respond…

Sun Microsystems inc is now in real trouble - the company’s stock price continues to move downward at a seemingly ever-increasing rate.   It’s a shame - the company has amazing assets that, in the long term, could generate massive share holder value.   The company also has, thanks to CEO Jonathan Schwartz, a fabulous long-term strategy. However, there is now a pressing short-term problem that needs to be addressed if the company is to survive. The number one problem for Sun is the sales and marketing of its hardware. Quarter after quarter after quarter, the company has failed to increase its top line.  Despite the big problems with the global economy, the market for server hardware is plenty big enough for Sun to significantly increase its sales.  It’s now time to fix this problem.  Herewith, my 30-day plan for starting to increase the company’s top line.

Firstly, why isn’t Sun increasing its revenues?  The answer to that is easy - Sun’s sales team is a walking, talking disaster.   It will take time to turn that around, even assuming they have a credible plan of action (which they may, or may not have - the company is reported to be in the process of firing 60%-70% of its sales and marketing people - that could well be the right move, but the devil is in the detail on that).   I don’t believe the company now has any time left to fix their sales team.  So, what to do?  Here’s what I think should be the top three priorities for Sun Microsystems.

1. Offer Windows Server 2003 and 2008 out-of-the-box on Sun x64 hardware

Windows Server is probably 90% of the market for x64 servers.  Currently, Sun is all but ceding 90% of the x64 server market to its competitors.  Time to change that. Let customers buy a box from Sun with Windows Server (either 2003 or 2008) pre-installed and working.  Sun hardware is already certified to run Windows Server, and already has an OEM agreement with Microsoft in place.  So, it’s not difficult to make this change happen. And, by the way, don’t just allow customers to buy these products, push them enthusiastically.  Bring this new offering to market within seven days.

2.  Focus On Direct Selling From the Sun.com Web-Site

Acknowledge Sun’s sales force is largely incompetent; especially when it comes to dealing with smaller customers.   This can’t be fixed quickly - the problems run deep and are embedded within the culture and processes of the company.  So,   side-step these problem by having a focusing on direct sales from the Sun web-site.

The current Sun on-line store is a disaster.   It looks like it was built using some crappy off-the-shelf e-shopping cart system. For any potential customer that doesn’t understand Sun’s product offering, it’s next to impossible to find products that are useful for them.  And actually, it’s pretty damn hard to find what you want, even if you do know what you’re looking for.

That means a new web-site is needed.  Don’t try to re-invent the wheel here.  Acknowledge the sales and marketing people don’t know what they’re doing, and simply copy the Dell.com site.  Is Dell.com as good as it could be?  No it’s not.  However, it’s several orders of magnitude better than the Sun.com site.   Also, copy Dell’s direct sales process with respect to account managers for repeat sales.  That means, reply to e-mail requests for quotes within two hours of receiving a request for a quote.

Bring the new web-site on-line for the x64 server range (with Windows given equal billing to Solaris) within 30 days. The rest of Sun’s offering should be be incorporated witin 90 days.  Where the purchasing decisions are complex, which they are for some hardware, there should be a way to e-mail a sales account manager to help the customer.

3. Clearly articulate a single unique selling point for Sun x64 Servers

Why should a customer buy Sun x64 servers vs those from HP or Dell?   There needs to be a single, easy-to-understand value proposition that is unique in the market place; and it needs to be compelling to customers.  For example, as long as it’s true (which it may or may not be), this USP could be: “Sun Microsystems x64 Windows, Linux and Solaris Servers - The Green Choice. Sun x64 servers use less power and emit less heat than equivalent competing offerings, and so save you money on your power and air conditioning bills.” Whatever it is, decide on this USP within 30 days, and articulate clearly and prominently it on Sun’s new web-site.

Bottom Line

Note that I am suggesting that these should be the top priorities.  That means, if resources need to be moved from other projects to make it happen - so be it.   For example, if top talent software engineering resources need to be drafted in from various projects around the company to make this happen, so be it.  Remember though, “NO CREATIVITY REQUIRED”.  The job is to make the new web-site work like Dell.com.  Not “like Dell.com rendered unusable, because someone thinks they can make the site better Dell.com.”

There’s lots to do after this, of course; particularly around leveraging the benefits of Sun’s software stack (xVM virtualization, Open Solaris [including ZFS, MySQL, Java etc). However, the above is what I think the 30-day plan should be.   After those 30-days, the priority should become getting xVM Server delivered to production, to enable Sun to deliver out-of-the-box virtualization solutions (including bundling OSes like Windows Server on top of it in pre-configured VMs) at world-beating prices.

Back in 1999, we contributed some software, written in Java, to a Free Open Source Software (FOSS) project.  At the time, it was a quite neat idea -  a software framework designed to enable developers to take any kind of file, in any format, and make that file appear to be an XML data stream, even though there’s no XML anywhere in sight.  Specifically, so-called SAX2 events are “automagically” generated by the framework, when non-XML files are input into the system.  Why is that a neat idea?   Well, because unlike most streams of data, these kinds of streams are self-describing.    That makes the life of a developer super-easy because it means that they don’t have to spend hours, days or weeks writing special code that “understands” how a particular type of file is structured. Rather, the software framework does the hard work, leaving the developer free to think about what they actually want to do with the data contained in those files.

That was 1999.  Today, it’s old code, and an old idea.  However, as is often the case,  software can live a long healthy life, way past the expected life the original authors had for it.  Indeed, that’s one of the main value propositions of FOSS - let the code be free, in the sense of freedom.   Why? Because, when code is free like this, it means other developers can make use of that code in whatever ways they like, even when the original authors cease to be involved with it.

Which brings me to today, when I received an e-mail from a Debian (Linux) developer.    A Debian package he’d put together, that uses our code, had been rejected by the Debian administrators due to a missing license information in one file.   We were meticulous in putting this information in the Java source code of our contributions.   However, it seems that a single XML DTD slipped through the net, and we hadn’t put any FOSS license information in that one file.   Getting a package rejected for something like this is obviously a bit of a pain for the developer involved.

So, the moral of this story is simple - if you’re developing FOSS, make sure you don’t forget the license headers in all your code! You never know when people will need that information in the future…

When Mike Arrington posted a story on Techcrunch today, saying that he has heard from multiple sources that Microsoft and Yahoo! are back in talks about a full acquisition, Yahoo!’s shareprice jumped.  According to Techcrunch, Microsoft is planning to offer something lower than than previous best offer of $33 per share.

What will that new offer be?  $30 per share?  $25 per share?   The best comment on this I’ve seen on this was in the comments section on Techcrunch. “ben”, at comment #15, said,

i heard the deal is now down from $33/share to 33 cases of zunes, an xbox, and a copy of guitar hero

That just about sums up the farce surrounding this deal in recent months…

Details of the O2 packages for iPhone 3G in the UK are starting to emerge.    The packages look fantastic compared to the previous deals.   Here are the details for the UK:

• Pay monthly
  • iPhone 3G 8GB will be FREE on £45 monthly tariff and above
  • iPhone 3G 8GB will be £99 on £30 and £35 tariffs
  • iPhone 3G 16GB will be FREE on £75 tariff
  • iPhone 3G 16GB will be £59 on £45 tariff
  • iPhone 3G 16FB will be $159 on £30 and £35 tariffs
• Pay As You Go
  • Details not announced yet, but iPhone is coming to Pay As You Go for the first time (however, see Comment #2 in the comments of this post)

At these price points (which include “unlimited data” - no confirmation yet on whether this is true unlimited, or subject to a fair use policy), I think there’s going to be huge demand in the UK market.  For anyone in the market for a “large form factor” phone, iPhone 3G is now the right phone to buy for probably 80% of customers.

I expect that Apple is now going to grow their sales to dominate this device category over the next five years. From what I’ve seen, other handset manufacturers simply aren’t going to compete effectively. Their actions suggest they have spectacularly failed to understand essentially all of the key issues that will make their offerings unattractive compared to Apple’s in the future.   To be honest, there should be senior executives getting fired left, right, and center in the mobile industry - this stuff really shouldn’t be beyond them. Whereas, Apple has demonstrated that they now understand just about all the issues that have been holding them back.  It didn’t take them long, either.  Just a few months in the market was all they needed to figure out how to evolve a winning strategy. If Apple can come up with a similarly spectacular offering for small form factor phones, they could rapidly dominate that market too.

Well, Steve Jobs and Apple delivered in terms of content: the 2008 WWDC keynote was significantly better than last year’s.  That’s the content, by the way; not the presentation itself, which seems not to have delighted everyone e.g. see the coverage at Engadget. (not enough Steve Jobs, I guess). Herewith, my initial reactions…

New iPhone 3G - Not Perfect, But Much Better!

• Apple’s offering is much improved. Now, the iPhone has 3G; and the business model is better, with a reduced up-front cost to get an iPhone in your pocket.
• Prices will start at $199 in the US (let’s see what the prices turn out to be in different countries).  iPhone 3G goes on-sale July 11, and I expect it to do significantly better in terms of sales than the previous model.

Large Numbers of iPhone 2 Software Early Adopters, Including The Enterprise

• The iPhone 2.0 SDK has been getting broad early support from developers. There are at least tens of thousands of developers working with iPhone (and it may be more than that).
• Significant numbers of corporations (especially larger corporations) have been exploring using iPhone in places where they currently use Blackberry.  That say’s they’re open to the idea of switching from Blackberry to iPhone.

Games On iPhone Will Be Super Popular

• The graphics capapability of iPhone (not least, the high-res screen) beats other mobile handsets by some margin, and the devices’s built-in sensors of lots of potential for interesting game control (like the Wii)
• Prices will be sensible - both Sega and Pagea Software will each be charging $9.99 per game
• Taken together, this combination of sensible pricing, unique graphics quality, and uniquene game play, means that games (and other entertainment software) on iPhone are going to be big.

There will be lots of other great software too (and the push notification service Apple announced today will enable new classes of app to be developed); and the Enterprise and new Ad Hoc distribution methods look good. The iPhone 2 software update will be pushed to all iPhone owners in early July, which is when the App Store will be open for business.

Bottom Line

• In terms of market share, the iPhone is now well-positioned to overtake both RIM’s Blackberry devices, and all Windows Mobile devices combined, by the end of 2009.

Literally, just a couple of minutes ago, during Steve Jobs’s 2008 WWDC keynote, Apple announced two new brands: MobileMe and Me.com

Now, we’ve been dropping clues to the name of one of our forthcoming brands for some months now (for example, see the photos on www.psynixis.com ).   Given Apple’s announcement, though, I wanted to get our trademark out there in the open early, in case we decide not to change it (although, to be honest, I think we probably need a new name now)… I’m not saying what this mark relates to precisely, other than to say it’s for an offering that’s at the interface of mobile devices and the web.

In a few hours, Steve Jobs will deliver the keynote presentation at Apple’s 2008 World Wide Developer Conference (WWDC.   I hoping (and expecting) it will be much better than last year’s, which I found pretty underwhelming.   Why? Well, at the 2007 WWDC, Apple announced no new hardware; and all that emerged was:

• The fact that the much anticipated “top secret” new features in Leopard didn’t exist
• There would be a release of Apple’s Safari web browser for Windows
• A statement that there would be no native third-party SDK for iPhone

At the time, I wrote:

Apple’s answer to the question of third-party app support on iPhone is the web browser i.e. AJAX, no SDK required… this simply isn’t good enough as a third-party application strategy.

It seems that Apple agreed with me because, in March of this year, the company announced a roadmap for a great SDK, along with a decent commercial strategy.   So, what will happen at this year’s keynote?  In a couple of hours, we’ll know…