Related link: https://www.sdforum.org/p/calEvent.asp?CID=990

I’m very proud to be one of the people organizing SDForum’s series of talks on The Future of Software Development. It’s an ongoing series, where leaders in our industry grapple with the question: “What’s the future going to be like?”

This month, on February 13, Stephen Wolfram is in the house. Here’s the abstract:

For three centuries scientists have looked to mathematics to understand how nature works - and to create the foundations for technology. But based on his surprising discoveries about simple computer programs, Stephen Wolfram, creator of Mathematica and author of A New Kind of Science , has developed a radically new approach.

Join us for an evening with Stephen Wolfram as he not only addresses many fundamental questions about science and the universe, but also suggests major new directions for technology.

Related link: https://www.raelity.org/archives/2003/01/27#document

Rael blogs rumors about Document, a supposed Word competitor from Apple. (What will they call their Office suite? I nominate “iRise.” Steve, think of the commercials! “And still I rise…”) Here’s a wishlist item in case of accuracy in these reports.

I think that programmers benefit from source control systems in ways that many other people might also enjoy — if only they knew how. Other forms of document change control certainly exist, but are terrible, for the most part; that is, they lack in features or robustness what CVS and friends lack in usability. Word’s “Track Changes” feature is Microsoft’s attempt; NetApp has another take on similar ideas.

It would be great if Apple (or some Office suite vendor) produced a Word-alike with a “Super Redline” feature, perhaps built on CVS or something else (consistent with Apple’s recent open source efforts — why is it that only Apple, of all the large computer companies, has figured out the real benefit of open source to their business? Everyone knows Steve and everyone watches what he does — why are their pattern matchers so terrible?). Let me compare my Document document (? blech) to your Word document and give me a good merge interface, a way to roll back to previous versions, branching, renames, and all the rest. For now, treat each user’s hard drive as their own personal repository; a networked Xserve version would come later.

(This idea grew out of some conversations I had with Brian about creating an open source repository for filmmakers — let me check in Quicktime files, sound effects, costume designs, edit decision lists, and whatever else, and let the rest of my production team receive diffs, have access to a current “build” of the film…you get the idea. This is one of the ideas I file under seductive but dangerous — I love it too much to evaluate it well. Filmmakers talk about “hanging a shot on the wall” (of the editing room) if you can tell you’re keeping it in the film because you love it on its own and not because it belongs. Brian rightly encouraged me to think of this idea as one for the wall. Here’s hoping the right movie for it comes along some day.)

As the economy and stock market continue to struggle for strong support and positive direction, look for some major Java camp players to make significant moves. Most of which will establish the trend of hardware vendors acquiring the struggling or, at least, challenged middleware/software vendors. Therefore, here are five potential Java industry mergers or acquisitions to watch for in 2003 and 2004:

1. HP Acquiring BEA Systems

Hewlett-Packard is different from other enterprise server players like IBM and Sun. While IBM and Sun offer complete enterprise middleware/software solutions (Web app server, Java/XML/Web services IDE, DB) for its server customers, HP really does not, unless you count the now defunct Bluestone. Recently, HP partnered with BEA and Oracle to offer various middleware/software solutions with its servers. Some analysts and others have considered a Sun acquisition of BEA. That’s not likely given Sun’s poor finanical leverage in terms of cash and/or current market value in the equity market. Maybe, a merger is possible where Sun would stengthen its middleware offerings. However, HP will acquire BEA first if necessary before they would let Sun do anything.

2. IBM Acquiring Novell

IBM may acquire Novell. IBM’s past ties to Novell in networking and as a recent distribution channel of WebSphere makes it the logical choice, especially as Novell augments its software offerings with new Java and Web services tools. Novell on its own will not make headway into the market, but it certainly offers potential ways to augment IBM’s WebSphere solutions.

3. Dell Acquiring Sun

Looking at Dell’s cash and strong equity market pricing per share versus Sun’s lack of cash and relatively poor equity market pricing per share is an invitation to Dell to possibly acquire Sun Microsystems. In addition to Dell’s financial leverage, Dell has been attempting to expand its market into the server-side arena. While Dell has made some in-roads into this arena, Sun already has established these ties with its current large business high capacity servers. Dell is also watching Sun’s success in the small business arena with its Cobalt Linux servers. Such an acquisition would give Dell the tools and market share to succeed in the server-side business market. It would also eliminate competition as well. Then Dell would only have to worry about HP and IBM. In regards to the effects on Java, it would be interesting to see what Dell will do.

4. HP Acquiring/Merging with Oracle

If HP is not able to acquire BEA, they may find a way to acquire or at least merge with Oracle. Again, HP would like to own and control middleware and software solutions for its server hardware. While I believe that BEA is more likely to be acquired, Oracle does offer a database solution that BEA does not.

5. Sun Merging with BEA or Oracle

While there is merit for Sun to merge with either BEA or Oracle, politics, ego and the above make this the least likely to happen.

Others to Watch

Borland, Macromedia, and Adobe are other significant market players to watch: Look for Borland to merge/be acquired by either a hardware or a middleware vendor (i.e., BEA Systems, HP, and/or Oracle). Macromedia and Adobe could merge with Apple long-term as well. Time will tell.

Place your bets here.

Related link: https://www.sdforum.org/p/calEvent.asp?CID=975

Richard Gabriel, distinguished engineer at SUN and author of Worse is Better, will be talking about the Future of Software Development this Thursday evening.

More precisely, In this talk, Gabriel will consider how new metaphors for software construction, drawn from areas such as biology, physics, and complexity science, can help us create more appropriate and robust systems.

Good stuff.

Related link: https://www.googlert.com/

While he wasn’t inspired by my call for Googlyeyes, a Google APIs application idea, Gideon Greenspan has done a great job of putting together the application I was wanting. Googlert takes your Google searches and re-runs them periodically, sending you the new results. This is great for applications like finding new articles about a company, or new Web sites about a topic of interest. Most excellent.

Related link: https://www.mdd.uscourts.gov/Opinions152/Opinions/sun_order0103.pdf

I could be wrong, but reading the order (PDF link) requiring Microsoft to ship Java with Windows, it looks like Microsoft has an out, of sorts:

Abandon the .NET Frameworks.

Paragraph 10(a) of the order defines the “Windows PC Operating System” (in part) as any Microsoft OS that “incorporates all or a portion of the .NET Framework, or for which Microsoft offers or distributes all or a portion of the .NET Framwork for automatic or optional installation.” Similar language defines Internet Explorer. So, if Microsoft drops the .NET Framework altogether — or perhaps, just makes this a technology users would need to buy as a separate product or install just as Java is installed today — they can go on shipping Windows (which would then fall outside of this definition) and MSIE without Java and prevent Sun from making any gains through this order. If you take the more open interpretation, Microsoft could sell .NET as part of, say, Visual Studio.NET and allow developers to ship a runtime with applications (just as is done for Java applications today), and not have to ship Java with Windows or MSIE.

Obviously, this seems like a very unlikely scenario — first, Microsoft will appeal this order and I suspect they will win on appeal; and second, Microsoft has invested an enormous amount in the .NET Framework, and without .NET they have no immediate answer to Java’s success with developers. They could reasonably judge that this would create a greater long-term harm for them.

On the other hand, some reports on this order have suggested that this is an enormous victory for Sun, one that will enable them to ride Microsoft’s distribution to great success. If Microsoft were to lose its appeals and did believe Sun stood to gain an enormous territorial advantage through this order, maybe their long-term position would be furthered by making this battle a stalemate. I find it interesting that the order gives them that option; it should not have, and Sun’s lawyers should have argued (or should have argued more strenuously) against any back door exit.

Related link: https://www.macworldexpo.com

At last week’s Macworld Conference and Expo in San Francisco, Sun wasn’t very visible. That’s surprising considering that Apple is very important to Sun’s plans on getting Java accepted on the desktop and much more. Furthermore, I believe that given both companies’ poor financial positioning in the equity market, an Apple-Sun merger or, at least, partnership would make sense for the following reasons.

Apple Takes Java Further.

Apple is currently the leading desktop implementor of Sun’s Java programming language, specifically the Java 2 Standard Edition platform, which includes the desktop JDK and the Java Foundation Classes (JFC), including Swing and Java 2D APIs for application GUI development. Additionally, Apple could potentially take the Java Media APIs further as well. Since Mac OS X is based on an open source FreeBSD kernel, Sun is hoping other open source advocate firms such as Red Hat adopt Java as these firms make headway in the desktop market.

Distribution Channel for StarOffice?

Sun may view Apple as a potential primary distributor of the StarOffice suite. Apple may be considering a suite of office software that replaces MS Office. There is speculation that Apple may adopt StarOffice and include Keynote presentation and other recently announced software, and distribute/bundle with certain Mac units.

Enterprise Mac OS X?

Apple may view Sun as a key to expand beyond the desktop and multimedia hardware markets; and gain legitimate entry into the enterprise arena. Sun’s server hardware and its market may be attractive to Apple. Apple would like to get its Mac OS X for the rack-mount server out there as an accepted server-side, enterprise OS solution. What better way than Sun?

For Sun, they may see this as a new selling opportunity for its Sun ONE Web application server suite of middleware and software. Although Macromedia’s JRun could be some competition, my hedge is that Apple would abandon its current emphasis on WebObjects in favor of a Sun merger or partnership here.

What do you think of such a potential merger or partnership?

Related link: https://news.com.com/2100-1001-980886.html?tag=fd_top

According to CNET News.com: “A federal judge on Wednesday ordered Microsoft to begin shipping Sun Microsystems’ Java with the Windows operating system within 120 days, after the companies fought over implementing a ruling he made last month.”

If this isn’t proof that developers should be open to developing both Java and .NET-based applications for Windows, Linux, BSD and other OS, then I don’t know what is.

Related link: https://www.bitwaste.com/wasted-bits/archives/000398.html#000398

Raffi would like to be bitten by Apple (or, at least, one of its products). How could Apple keep the developers streaming towards its OS X platform happier when the products they buy are surpassed?

Apple’s proclivity for splashy new product announcements tends to leave people who have invested in their products feeling left in the dust, far too often. Raffi was complaining about the new Airport Extreme cards not working in his not-yet-six-month-old TiBook. Here’s my reply:

Are you on a network that gets better than 11Mbps (that is, are you on an T3 or better)? I imagine at MIT you are but I imagine that the vast majority of tibook owners have your home setup (cable/dsl) or a work T1
and that’s it. Even if you have that much pipe, could you use it? I bet MIT’s network has some rate limiter somewhere that would prevent you from getting over 11Mbps even at midnight on Christmas when the network is dead.

I agree with your post in general — I think it’s totally lame that Apple screws people who just bought tibooks for Christmas or whatever. I think they should have an upgrade path where you could send in your old
tibook and they would upgrade some of the pieces (like say the superdrive) with whatever the latest and greatest is. Even if you didn’t get everything at least you would feel like they remembered that you bought a big product from them a month ago or whatever. I bet people would even pay for the right to have their tibooks upgraded if the price was reasonable.

But on this particular issue I think it’s not worth getting excited about. Bitch about the superdrive and the wifi antenna on the screen! Those are things that would directly improve your tibook ride.

That’s my two bits.

Apple is now attracting a different kind of user (although, it’s been a problem for years — check out this Doonesbury cartoon from May of 1994 — that’s a Mac all right). The mass of developers swarming all over new Apple laptops and OS X won’t be happy unless they have the latest and greatest — and Apple is kidding themselves if they think that means people will buy a new laptop once a year. I think they need to figure out how to tend to the developers who are buying their products. Happy developers == more development on your platform == more customers.

One of the companies that sells the primordial iPod ancestor, the Compaq Personal JukeBox (PJB), has a great program where you can upgrade your existing PJB for a fee. You send them your PJB via FedEx, and they:

1. Install a new Toshiba 40 GB Hard Drive to in your PJB 100
2. Transfer all of your existing music to the new 40 GB drive.
3. Run system diagnostics and reassemble the product.
4. Upgrade to the latest software.

(Taken from their upgrade offers page.) Then they FedEx it back to you. Most customers report a one-day turnaround. People are ecstatic to be able to renew their investment in the product, and it makes users much more loyal to and appreciative of it. Apple needs to figure this out. They even have a newly-built advantage: Apple Stores that could do upgrades on the spot.

I know that there’s an Apple operations executive having convulsions just from me typing these words — either from fright or more likely from hilarity. Yeah, there were probably only a few thousand PJB owners, and the upgrades probably came in around one a day at peak. But come on, Apple, figure it out — it’s not like you have that many customers yet. Certainly not so many that you can afford to lose a customer like Raffi.

Related link: https://www.bayarea.com/mld/mercurynews/business/technology/4912773.htm

According to the San Jose Mercury News on Saturday, Microsoft is apparently planning to “reel” in its .Net tag brand. Or is it .NET or .net? Anyway, that’s probably best. According to this article, they’ll eliminate the brand on the Windows OS. Applications will surely follow. Then according to this article, they’ll use .Net in the following way:

“… identifying products and services that support Web services interoperability with a ‘.Net Connected’ logo…” This likely refers to development languages and tools like ASP.NET and Visual Studio .NET.

Related link: https://www.borland.com/optimizeit

I’ve been evaluating the latest release of Borland OptimizeIt over the past few days, and must say that I really like it. It helped us track down some significant memory leaks and performance issues in our application after the first few hours of use, so it is definitely worth the money. We have really struggled with memory consumption in our Java client, but now we think we can fix the issues.

The only drawback I’ve noticed (so far) is that they don’t list JBoss as a supported app server. They do provide instructions for hooking up to JBoss, but it requires a little more work than the “pre configured” app server support. Overall, I’m very satisfied with OptimizeIt.

Related link: https://www.idg.net/ic_1008496_6134_1-3523.html

At the San Francisco Macworld Expo, enterprise Java for the Mac OS X does exist as evidenced by this Macromedia announcement. Yesterday, Macromedia announced the availability of its JRun J2EE container or Web application server for the Mac OS X.

According to InfoWorld: “Priced at $899 per processor, Macromedia’s JRun 4 for Mac OS X is a commercial J2EE-compatible application server used at more than 10,000 companies worldwide, according to Macromedia. JRun could be used on Xserve, Apple’s rack server for Mac OS X, according to Macromedia.”

Therefore, as a result, enterprise Java development and deployment options for the Mac OS X rack server now includes the following noteworthy software (not comprehensive):

J2EE Web Application Server/Container

• Macromedia JRun 4 for Mac OS X
• Apple WebObjects

Java IDE with J2EE Support

• Borland JBuilder for Mac OS X
• Oracle JDeveloper for Mac OS X

Yes, Oracle also just announced the JDeveloper for Mac OS X. Perhaps, Oracle also plans on offering its database and other solutions for the Mac soon. Of course, it depends on how much the Mac OS X is accepted as a reliable and value-added enterprise solution.

While I have my doubts about enterprise Java for the Mac, I think the jury is still out on this. What do you think?

Related link: https://www.neward.net/ted/weblog?date=20030107

Recently, a number of authors and writers have been talking about AOP (Aspect-Oriented Programming), and how incredibly powerful and wonderful the whole thing is. And yet, for the vast majority of them, what they’re really referring to is an old pattern called Interception. These people aren’t stupid, nor are they misguided; while Interception is a powerful mechanism in its own right, it’s not the same as AOP, and while they do share a number of defining characteristics, to understand AOP as Interception is like thinking OOP is data structs plus a bunch of function pointers. Just as we didn’t understand objects until we got past the idea that objects are “just code and data”, if we’re to truly understand the potential power inherent in AOP, we need to get beyond this thinking that Interception and AOP are the same thing. They’re obviously related at some points, but that’s not the same thing as equality.

Anyone involved with securing systems now days is likely to become
rather too involved with only part of the problem: Prevention.  In Bruce Schneier’s Secrets and Lies,
he stresses the importance of “the rest” of the analysis: Detection and
Response, which together with Prevention form the synergistic triad of
security.

Bruce can take heart: a recent Santa Fe Institute workshop entitled “Resilient & Adaptive Defense of Computing Networks” is setting the stage for a different approach to security, one modeled on natural resilience often seen in nature.  These techniques
are adaptive: they respond in natural ways to the behavior of the
system.  One holds opinion polls amongst the participating cache
servers to agree/disagree on the integrity of the data they
hold.  Another looks at the packet traffic within a network,
looking for signatures of “normal” use and responds when abnormal
behavior is seen.

Another interesting theme is that several techniques can be used
together, one protecting from virus attacks, another from break-ins, a
third on data integrity, another checking the subnet health.  The
Whole is greater than the Sum of the Parts in these situations.
 By their simplicity and independence, these adaptive approaches
avoid the brittleness typical of Prevention-only systems. Much of this
work originated with Stephanie Forrest’s ground-breaking “Computer Immunology” work at University of New Mexico.

Let me give one concrete example
(click on Acrobat logo for .pdf file) to illustrate these
approaches.  This is from Matt Williamson, of HP Research Labs,
and a earlier a student of Stephanie’s. It is based on the observation
that systems tend to limit the number of hosts they talk to at any
given moment.  Matt keeps a short list (5 is common) of “active”
hosts that get full response by the computer.  New hosts are put
into a queue that is slightly delayed, typically by a second.  As
old active hosts age, they are replaced by the new nodes which now
operate at full speed.

This approach works quite well, tolerating “false positives” yet
effectively throttling viruses.  And it and others like it are
getting interested coverage in the media.

Generally, these systems have in common the idea of the computer
monitoring its environment, and learning what is normal behavior for
that system.  By carefully allowing new behavior to be first
checked, then adapted to, false positives are made benign.

Robert Ghanea-Hercock
(click on “people”), who holds these workshops at SFI, notes that the
next workshop will be November 5-6 2003, just preceding SFI’s Annual
Business Meeting.

What’s your take on moving beyond crypto?

Related link: https://www.macworldexpo.com/macworld2003/V33/conference/track.cvn?id=6&tID=24

Is there Java at this Macworld Expo? Yes, there is, but it’s really only one session in the Developer track.

Daniel Steinberg teaches this session as described on Macworldexpo.com: “Using Mac OS X applications is great but it’s only part of the experience. If you’d like to write your own Mac OS X application without having to learn about the Cocoa and Carbon APIs, then Java is for you. With the Java Virtual Machine included in every version of Mac OS X, you have the perfect environment for playing around with writing programs for your Mac. In this session, you’ll look at the free tools available for you to develop Jaguar applications in the Java programming language. You’ll see how to write a quick visual program and how to package it up to distribute a double clickable application. You’ll be provided with links to resources for more information. This session provides an excellent introduction to writing in Java on Mac OS X and important steps that need to be taken to customize your Java apps for the platform.”

Other than this, I would look for any session that covers or includes Mac OS X or the more recent Jaguar with Apache Jakarta Tomcat. Furthermore, there wasn’t much on Cocoa and other programming either. While I realize that the Macworld Expo and the Mac audience in general is geared toward the animator, artist, editor, publisher, and writer, I expected a larger developer track than just three courses with only one on Java. This is somewhat surprising given the interest in Java for the Mac.

This is somewhat surprising given the interest in Java for the Mac. What do you think of the lack in Java coverage?

Related link: https://news.com.com/2100-1040-979461.html?tag=fd_lede1_hed

Macworld Expo: Jobs offers Keynote. No, I’m not talking about the keynote speech, but Keynote, the application. In this News.com report covering the Macworld Expo in San Francisco today, Apple Computer CEO Steve Jobs talks about his company’s latest offerings, including a new presentation application, called Keynote.

“‘We built this for me, and so I wanted to share it with you,’ Jobs said, adding that he used test versions of the software for all his Macworld presentations in 2002. The program conceivably would compete with Microsoft’s version of PowerPoint for the Mac. ‘Keynote imports and exports PowerPoint,’ Jobs said. The program also supports Adobe’s PDF and Apple’s QuickTime formats. Keynote will be available starting Tuesday for $99.”

Given the end of the Apple-Microsoft software partnership agreement last year that dated back to 1997 as part of a “bailout” deal, Apple is now “free” to develop and sell applications like Keynote to its customers. Look for this to be the beginning of an Office-like bundle that could also include a word processor, spreadsheet, and email. I expect that Apple will likely develop internally, but may consider possible acquisition or use of Corel’s WordPerfect and Quattro Pro suite or even Sun’s StarOffice.

Also announced in today’s keynote: Safari. No, it’s not the O’Reilly Network Safari bookshelf, but Safari, Apple’s new turbo-charged Web browser to replace Apple’s prior use of IE. Apple’s Safari should suit the multimedia user.

This may leave Microsoft out in the cold when it comes to Apple. However, Microsoft is still a factor in the Apple universe, and .NET may be the key. It’s no coincidence that Microsoft’s Rotor project allows .NET applications to be developed and used on FreeBSD based systems and that Mac OS X is based on a FreeBSD-like kernel.

Related link: https://www.sdforum.org/p/calEvent.asp?CID=983&mo=1&yr=2003

The January 14 Meeting of the Emerging Technology SIG will
feature a talk by Chris Thomas, Chief E-Strategist of Intel.
Chris will be speaking about the current state of the enterprise software industry (from Intel’s point of view) and the trends that Intel thinks are important over the next 2 to 4 years.”

Please note that it will be held in room H-1 of the Cubberley Community Center, in Mountain View. Cubberley H-1 is a much larger and nicer room than the one we were using at the Mountain View Community Center. We’ll be meeting at Cubberley H-1 from now on.

More details, and the full abstract of the are available here.

We’ll also be raffling off three copies of O’Reilly and
Associates’s “Designing Embedded Hardware” and a single
copy of Steven Wolfram’s “A New Kind of Science.”

We’re currently planning something like:

     7:00 Pizza, networking, and small-talk
     7:20 Chris begins speaking

As always, events are free to SDForum members and $15 to everyone else.

Hope to see you there,

William Grosso

Co-chair, Emerging Technology SIG

SDForum: the emerging business and technology connection

Since I posted my
predictions for 2003, I’ve received a fair amount of feedback. Most of
it has centered around the third prediction:

3. $300

Linux boxes at Walmart (and other low-end retailers) will become the standard system for

home users. I honestly think that the introduction of these boxes by mass-market retailers was

was the most significant tech news for 2002, and that the significance was mostly missed.

Walmart is

href="https://www.walmart.com/catalog/content_shops.gsp?cat=102317&path;=0%3A3944%3A3951%3A10231

7">lowering costs by using Linux. And they’re producing PCs which are more than good

enough for most applications. This is huge, and along with Apple’s resurgence, signals the

re-emergence of Java on the client side.

The objections seem to fall into two main categories:

• Sure, some people will buy those boxes initially. But sales will fall off because Linux is not usable enough. Where “usable” is a combination of “user friendly” and “runs the right software.”
• Sure, some people buy those boxes. But they install (pirated copies of) Windows on them soon after.

It’s not very difficult to address these objections. There are lots of responses that could be made. For example:

• There is enough software for most tasks. For niche domains, or for power users, Windows retains an advantage. And for experienced users, Windows has a significant advantage (If someone knows how to use Word already, the cost of using a different word processor has to include the cost of retraining them). But, in general, for low-end use (simple word processing, e-mail and web-browsing), Linux has pulled even.
• People are happy with
  Linux. It hasn’t been exhaustively validated yet, and there are no statistical surveys, but the focus on usability seems to be paying off (see, for example, href="https://www.distrowatch.com/">these reviews).

• People don’t actually change their software very often, let alone install a new operating system. Consider, for example, Internet
  Explorer. href="https://hotwired.lycos.com/webmonkey/01/30/index3a.html?tw=authoring">Internet Explorer 6 was released in 2001. And yet, 45
  % of web traffic is still using Internet Explorer 5. And this is in an environment where the various security holes and warnings have actively encouraged upgrading.

• It’s much harder to pirate Windows than in years past. While href="https://www.newsfactor.com/perl/story/14725.html">it’s
  not foolproof, the fact is that the barrier to piracy has gone up substantially.

In short: there’s enough software, it’s usable enough, people don’t like changing software (much less changing operating systems), and pirating (or even just installing) Windows is a substantial technical hurdle.

But I don’t want to spend too much time on those arguments. Because they’re “here and now” arguments that talk about “what are people in January of 2002 doing with their Linux boxes from Walmart.” And I want to talk about the future. What I wrote was a prediction about what’s going to be true in December 2003, not a statement about what’s true now.

And there are a few interesting points which I didn’t mention in the predictions column, but which I want to bring
up here. None of these are killer arguments, but they cumulatively imply low-end Linux is awesome.

There’s a lot of vendors moving in. There are literally dozens of vendors selling Linux desktop boxes. It’s not just Microtel selling boxes at Walmart. There’s a lot of movement here, and there’s a lot of competition. That means a fair number of companies will die. But it also means that the products will quickly evolve, and that’s good for the customers.

These are not the networked computer, 2003 version.. That was a corporation-only vision, and required massive buy-in (for example, IBM’s network computer required an IBM server and only interoperated with IBM equipment). These are all full-fledged independent systems based on open standards. People can buy and use single units and those single units will work as standalone units and as part of existing networks. As part of this, there’s a very nice incremental adoption path available for corporations (it’s not an all-or-nothing decision).

China and India. They’re big. More than a billion people each. And they’re just at the beginning of the computer adoption curve. But it’s beginning to look a lot like both of these nations are going the linux route. For example, here’s href="https://www.techtv.com/screensavers/linux/story/0,24330,3395670,00.html">an article on China with some pretty good links.

On the other hand, it also looks a little like Chinese programmers might be forking linux without

sharing code. That’s interesting– it’s hard to fight piracy across international lines. But it might be equally hard to enforce open-source licenses.

Does Microsoft have any brand loyalty?
One of the ways you fight encroachment from
underneath is by having brand loyalty. I buy Del Monte tomatoes rather than the supermarket-brand tomatoes because I know Del Monte does canned tomatoes well. And I’m more
likely to buy other Del Monte vegetables as well, because I trust Del Monte to do a good job. I have never heard anyone say anything even remotely similar about Microsoft or Microsoft products.

What would Michael do? I’m not a religious guy; this isn’t about the Archangel Michael. It’s about Michael
Dell. And the question is simple: if he were an undergraduate at UT Austin today, where would he be focusing his energies? Where are the entrepreneurial opportunities for ambitious and bright entrepreneurs who have a limited budget? I think Linux has it all over Windows
here, and that means that there’s going to be more market innovation and responsiveness from
Linux PC makers over the next 12 months than from Windows PC makers. The people who make Linux PCs are just going to out-innovate everyone else in the mad-scramble to move some boxes.

It’s got the right spiral. Suppose you buy some of these arguments, but not all of them. And you say “All right. I’ll grant you that Linux desktops will get maybe 10% of the new purchases on the low end of the market by years end. Windows will still be dominant. New machines will still mostly be Windows. And all the older machines will be Windows.” That’s a reasonable proposition to make. Then I have one question: what happens in 2004?

So what do you think? And where does all this lead?

My wife made an interesting comment the other day. Something like “gee, honey, you are really down on technology lately.”. This was after I was cursing at the computer because yet another program was broken.

Here are some recent examples:

• I tried to download some evaluation software from Borland the other day. After filling out their little survey and going to the downloads page, I find that the link is broken. They swear that it works, so I tried it from home. It still doesn’t work. The URL looks a little weird, so my hunch is that they are pointing to an internal machine that is not available from outside of their network.
• I tried to take advantage of Missouri’s great new program where you can pay for updated license plates online. The problem is, the system fails every time I try to submit my personal property tax info. I guess it’s back to paper.
• I tried to register a new account with a service called Privista that lets me track my credit report online. The registration totally bombs when I enter my credit card info, however, printing a huge JSP stack trace. So now I’m wondering…did they bill my card? Do they have other nasty bugs that could expose my data to other surfers?
• I tried to download updated JDBC drivers from Oracle. When I try to create an account for myself, their web site craps out with a big stack trace.

The list goes on and on….NOTHING seems to work these days! OK, that’s an exaggeration, but the percentage of application failures is very, very high. My examples above are all web apps, but the same holds true for all other types of applications.

On a positive note, I can’t imagine that I’ll be out of work any time soon. There are too many bugs to fix!

Domain Madness

OK, I admit it, I may have too much time on my hands,
but I was looking for a new DNS domain name, and decided to shove all of
the 234,937 words in /usr/share/dict/words through Whois, collecting those
not having a .com entry. The script is attached below. Currently the words
file lists the words in Webster’s Second International, who’s 1937 copyright
has expired.

Oddly enough, the result is that 63% of the words are NOT
registered names!! That’s right, 147,886 words are not taken. That’s the
good news. The bad news is that many of them are pretty weird. I’ve stashed
these guys, both compressed and clear, on
   
https://backspaces.net/files/NonDNSWords


   
https://backspaces.net/files/NonDNSWords.gz

For example, here’s the list of all 43 4-letter words not taken:

grep ^....$ NonDNSWords
bikh    fowk    hawm    koae    odso    shlu    waup    yeuk    yirndird    frib    hewt    kuar    oime    suld    wusp    yigh    yirrdowf    gawm    jaob    mowt    paut    syrt    wype    yilt    yuftdowp    ghuz    jewy    munj    phoh    uily    yalb    yirkemyd    gype    jhow    niog    rynt    wauf    ycie    yirm

I can’t see one that calls out to me, really. A lot of these do not appear in my dictionary, but the Second International was known for stretching!

Here’s a random sampling of 100 of the 6 letter critters:

grep ^......$ NonDNSWords | ran 100 | column -xdiaene evener tummer burdie palpon coccid taxwax chanst madefy buntalhaggly masted untone dutied unmiry cynips psetta otitic gawcie beflagmidpit orgyia tutory amylic begnaw punlet adigei scrank bedrop lusorydorize repale unmold snurly scotic unsing uplead hemine unnose stibicfunori cobcab yengee cahita rutuli menkib uptend sassak beflap crantsocyroe rugose avowry mogdad coecal elleck ptotic kommos amusgo lemosiavitic amorua cacara ideist reswim napaea reshut egeran lechea embolykorait uplick baeria kurvey ureido tuchit beroll adroop degged twiselkechel solate unbare hardim upwaft sullan tineal uramil ovinia pappoxforrad jacami unlean byrlaw thymyl scrobe lyncid crenic bepity anoine

..where “ran” is a simple awk script, below, to randomly select n lines
from a file. Its kinda spooky doing all this on Mac OS X .. it really IS
Unix.

Again, not a lot of love. By the way, there were 5,166 6-letter
words, so I likely have not shown some real winners in this sampling. Let
me know if you find some real winners.

This got me a bit curious .. how do the words work out by size?
I.e. how many words are 6 letters long etc? Time for another script, also
attached below:

/usr/share/dict/words           NonDNSWords 1         52     0.0221        . 2        155     0.0660        . 3       1351     0.5750        . 4       5110     2.1751         4         43     0.0291 5       9987     4.2509         5       1219     0.8243 6      17477     7.4390         6       5166     3.4932 7      23734    10.1023         7      10725     7.2522 8      29926    12.7379         8      16593    11.2201 9      32380    13.7824         9      20861    14.106110      30867    13.1384        10      22254    15.048111      26011    11.0715        11      20415    13.804612      20460     8.7087        12      17065    11.539313      14937     6.3579        13      12935     8.746614       9763     4.1556        14       8811     5.958015       5924     2.5215        15       5433     3.673816       3377     1.4374        16       3146     2.127317       1813     0.7717        17       1681     1.136718        842     0.3584        18        804     0.543719        428     0.1822        19        408     0.275920        198     0.0843        20        189     0.127821         82     0.0349        21         79     0.053422         41     0.0175        22         38     0.025723         17     0.0072        23         16     0.010824          5     0.0021        24          5     0.0034

Well, the most populous part of NonDNSWords is 10; here’s a sample:

dramseller floriation tractional clanswoman periphrasecyrtometer symphytize convolvuli mucigenous clamminesshyperacute myrtlelike unharbored ergonovine undertideddigressory preclosure parnassism habilatory boycottismnilometric paralgesic trimacular annelidian breezinessprelegatee admiringly scatophagy bonebinder morphinismendosteoma ranivorous undistinct solenodont scathinglyunfreckled unpanelled impalpably unemphatic staverwortgradientia cystospasm xenocratic cogredient rubescenceneurolytic unrebutted saponacity brachyoura depatriate

OK, I know you want to know the 5 24-letter words, so here they are:

formaldehydesulphoxylate
pathologicopsychological
scientificophilosophical
tetraiodophenolphthalein
thyroparathyroidectomize

..and, yup, antidisestablishmentarianism wasn’t there.

This did make it easy to search for substrings of interest. For example, I wanted to
find all the words with “plex” in them. There were 54:

grep plex NonDNSWords | column -x
amplexation     amplexicaudate  amplexicaul     amplexicauline  amplexifoliate
autocomplexes   cerviciplex     complexedness   complexionably  complexional
complexionally  complexioned    complexionless  complexively    complexly
decemplex       diaplexal       diaplexus       epiplexis       euplexoptera
ganglioplexus   holoplexia      intercomplexity kataplexy       myelapoplexy
nulliplex       overcomplex     overcomplexity  perplexable     perplexedly
perplexedness   perplexingly    perplexment     phantoplex      plexicose
pleximeter      pleximetric     plexodont       plexometer      plexure
pseudoapoplexy  reperplex       retroplexed     semiamplexicaul semiduplex
sextuplex       simplexed       supercomplex    triplexity      ultracomplex
unimultiplex    unperplexed     unperplexing    veniplex

This is a bit more interesting: holoplexia.com sounds nifty, as does nulliplex.com

So, I guess you’re wondering which one I took, right? Well, sadly,
none of them. While groveling around, I thought of a two-word critter I kinda
like: ComplexityWorks.com, so hmm..all this was a waste? I think not, but…

Scripts:

Check a list of words w/ whois.#!/bin/shpat=${1:-"^...*"}start=${2:-a}file=${3:-/usr/share/dict/words}words=`sed -n "/^$start/,\$p" $file | grep $pat -i`for w in $words ; do	whois $w.com | 	sed -n '/No match for/{s:.*for .::;s:......$::p;}' | 	tr A-Z a-zdone
For choosing N random samples from a stream:#!/bin/shsamples=$1awk -v samples=$samples '{a[NR]=$0} # Read in fileEND {        len=NR        for ( len=NR; samples > 0 && len > 0; samples--) {                i=int(rand()*(len+1))                print a[i]                delete a[i]                len--        }}'For sorting a stream by length:#!/bin/shawk '{a[length]++}END {	for (i in a) printf "%2i %10i %10.4f n", i, a[i], 100*a[i]/NR}' | sort -n

I’m curious: How did you pick *your* domain?!

Well, it’s that time of year again. Retrospectives and predictions abound; people are scurrying ’round talking about the events of 2002 and predicting what will happen in 2003. Since I’m all in favor of a little mad scurrying, I’m going to join in the fun.
I’m a little late out of the gate with this (What can I say? We started celebrating even earlier than Ted), but here are my top five technology predictions for 2003:

5. JBoss moves past 50% marketetshare in the EJB space. This is pretty much a no-brainer. JBoss is free, it works well, it’s got a large and growing number of users, and it’s just nice. I also don’t agree with Ted that EJB will be dead by year’s end, but that’s a different story entirely.

4. A massive shakeout in the CRM space. There’s just too many companies. And far too many of them are still venture-backed and looking for a profit. The combination of shaky finances and Microsoft looming on the horizon is going to cause buyers to pause. And that’s going to start the dominos falling.

3. $300 linux boxes at Walmart (and other low-end retailers) will become the standard system for home users. I honestly think that the introduction of these boxes by mass-market retailers was was the most significant tech news for 2002, and that the significance was mostly missed. Walmart is lowering costs by using linux. And they’re producing PCs which are more than good enough for most applications. This is huge, and along with Apple’s resurgence, signals the re-emergence of Java on the client side.

2. Broadband will hit 5 million new households a quarter. Right now, in the United States, broadband adoption is approximately 1.7 million households per quarter. And it’s growing at a rate of 9% per month. Since the report is from the third quarter, if we just assume the 9% figure holds steady, compounding the 9% monthly gets us to more than 3.5 million new broadband connections in Q4 of 2003. That’s a safe and reasonable prediction. But I’m going to go out on a limb and say that, by Q3 of 2003, as the United States firmly leaves this recession behind, broadband adoption will be skyrocketing.

1. The emergence of P2P strategy and simulation gaming networks. What happens when $300 linux-based servers(#3) can run free enterprise class servers (#5) from the basement using a broadband connection (#2)? And when a significant percentage of them are also running wi-fi networks? In the short run, I’m guessing one hell of a finely detailed version of Axis and Allies. In the long run? The rebirth of the internet.

And that’s it– my top 5 predictions for 2003. I’m finding it a little weird that I didn’t mention XML, web-services, Mozilla, wireless devices, or SDForum. Who would’ve predicted that?

Am I off base? What would you have included here?