Phil Windley's Technometria

« June 2004 | Main | August 2004 »

July 30, 2004

OSCON 2004: Open Source GIS with GRASS

When I was the CIO of Utah, I used to quip that every data record I cared about had a SSN or latlong coordinates. There's lots of things that are geographically categorizable, but commercial GIS software is expensive and consequently, geographic Internet applications are few and far between. Even things like MapQuest don't really do much besides give you directions. Why can't I annotate them, for example? GRASS is an open source GIS system that provides raster, topological vector, image processing, and graphics production functionality. I went to a talk today by SchuylerÊErle and RichÊGibson on using GRASS. Here's the slides. There are easier tools to use (like Quantum), but GRASS provides more advanced functionality using map algebra. That advanced functionality comes at a cost: GRASS isn't easy to use.

12:01 PM | Comments () | Recommend This | Print This

Poor Man's Streaming

I reported on Dana Moore's talk about using Jabber in highly available agent networks for DARPA a few days ago. I wasn't able to go to his and Rich Kilmer's talk yesterday on using Ruby as the control language, but Jon Udell did and offers a video interview of Dana and Rick after the talk. The interview itself is interesting in its own right, but the meta-story is interesting as well. Jon did the interview with an iSight and his TiBook. He then just uploaded the Quicktime file to a server. No streaming infrastructure needed. Most players start playing right away nowadays, so unless you're doing something live, you don't really need streaming. Creating something like this is relatively simple and is vastly under-used as far as I can see.

10:15 AM | Comments () | Recommend This | Print This

OSCON 2004: David Rumsey on Online Digital Libraries

This morning's keynote was David Rumsey, President of Cartography Associates. Rumsey's website is called he David Rumsey Map Collection and what a collection. The physical collection consists of over 150,000 maps, mainly from the 18th and 19th century. The online collection is no less phenomenal and is more than just a collections of GIFs or PDFs. The maps and other art is all interrelated and interactive. Its almost impossible to describe in words how incredible all this is. You literally have to see it to believe it. To get a feel, they have a Flash tour of the collection. The collection has a special browser that runs inside IE or Mozilla. This collection is a real high water mark for online digital libraries to strive for. I rarely see a keynote speaker at a conference get a standing ovation, but David did this morning. It was well deserved.

9:46 AM | Comments () | Recommend This | Print This

July 29, 2004

OSCON 2004: Dan Gillmor on We the Media

I wanted to go to Dan Gollmor's talk yesterday, but I there were 3 or 4 good talks going on then. I was glad to see Jeremy Zawadny's notes. This is the topic of Dan's new book, We the Media.

11:09 PM | Comments () | Recommend This | Print This

OSCON 2004: Ben Galbraith on Publishing a Medical Textbook with Apache FOP and XSL-FO

Here's the challenge:

• Entirely new kind of textbook.
  • Structured content, not prose
  • Extensive pictures
  • Books generated on demand
• Reuse content in other forms

The first attempt was the Microsoft tool approach using Word with special templates as an authoring tool, Word VBA to convert Word to HTML-ish format, access used to store content, and then VB and Framemaker macros would generate content. A whole generation of books was developed with this technology, but it was a mess and the content was not reusable.

The second attempt used a Java Swing-based editing tool (modified JTree with Word-like editing features and an XML binding layer). XSLT converted the XML into LaTeX and then MiKTeX rendered the LaTeX into PDF. The authors liked the authoring tool. The rendering system was too inflexible. LaTex is showing its age and the layout was unstable. Also the data binding was too inflexible requiring that it be recompiled with every change.

The third attempts was the replace the LaTeX stylesheet with XML-FO. This was more stable and flexible than the LaTeX solution.

The editing tool presents structured data as forms to be filled out rather than relying on authors (doctors) to write prose. Creating the editor was a big job, but worth the effort.

XML-FO v1.0 has some limitation (no support for multi-column region per page, etc.) but XML-FO 1.1 will eclipse TeX/LaTeX functionality. Apache FOP is an open source XML-FO processor. Its not fully compliant and may never be. The project was recently rebooted to start recoding from scratch. RenderX XEP is a commercial XML-FO processor.

One additional problem they ran into was the amount of computation that rendering a full medical text with all its images requires. Ben used JNGI to create a grid of computers to do the rendering a few pages at a time.

11:43 AM | Comments () | Recommend This | Print This

OSCON 2004: Freeman and George Dyson

This morning's key note was Freeman and George Dyson. Esther was supposed to be here as well, but she's stuck in Dallas. No wonder, I heard on CNN this morning that Dallas got 12 inches of rain in 3 hours. Yikes! The format was Tim O'Reilly moderating and asking questions of Freeman and George. Here are a few things that struck me as interesting:

With regards to possibly dangerous technical advances (specifically the topic was bioengineering), Freeman says there are three questions to ask:

1. Is it possible to put a stop to is?
2. Is is desirable to put a stop to it?
3. If you want to control it, what's the appropriate mechanism for doing that?

George makes talks about being a "maker." He says we went through a period where children didn't take apart things. They were too complicated and we've lost something in that. This is a great observation. My son just went to a "invention camp" at his school and took a bunch of stuff to take apart. I think it significantly changed his outlook on how things work and made him much more interested in how things work.

Freeman talks about the Achilles heel of the nuclear industry being that it doesn't work on a small scale. Open source software is about doing large scale things in a decentralized way. The small scale nature of any individual contribution is its strength.

Scientists are discoverers or explainers. Discoverers are makers and tool builders. There's an analogy about birds who look down from the sky and frogs who look up from the pond.

9:47 AM | Comments () | Recommend This | Print This

July 28, 2004

OSCON 2004: Groovy

Maybe the coolest thing I heard about today is a language called Groovy. Groovy is a dynamic scripting language that compiles to JVM bytecode. The result is that it can use the entire J2SE/J2EE API. That's leverage. Groovy would be a good candidate for a scripting language for integrating lots of Java code. The code is dense and fairly elegant.

5:41 PM | Comments () | Recommend This | Print This

OSCON 2004: Dana Moore on Jabber Messaging

Jabber is not just for IM anymore. Conversational interfaces are less structured, more flexible and can be ambiguous. They can support S2S (system to system) as well as they support P2P (person to person). Applications need presence, fault-tolerance, identity, and mobility the same way people do.

XMPP is the eXtensible Messaging and Presence Protocol, the basis for Jabber. Jabber streams XML messages. Conversations open and close with <stream> tags. The server works as a switch and doesn't maintain much state on users.

Pervasive services follow you wherever you go. The service knows that you're subscribed to them. Dana builds sensor systems and wraps Jabber communication code around them so they understand presence messages and can send status messages over XMPP. He describes an alarm system (briefly) built on XMPP. He uses the term "constellation of services."

The protocol is dirt simple. XMPP has three element types <presence>, <message> and <iq>. The <presence> and <message> elements are relatively self-explanatory. The <iq> message is for communicating with the server. The <iq> message, for example, can be used to tell the server that you have a service to offer. This allows services to be published to the server and other users to discover them.

Dana is working on the DARPA UltraLog project which is attempting to create extremely survivable software systems. The agent architecture for the system is built upon Jabber. The network is dynamically reconfigurable and can simulate attacks. This was made possible by using agile languages (Ruby) and Jabber. One benefit was that IM clients served as the ready-built user interface to the system.

Dana's bumper sticker summary of the talk is "Jabber enables Web services without the baggage."

12:16 PM | Comments () | Recommend This | Print This

OSCON 2004: O'Reilly Radar

Tim O'Reilly, in his annual O'Reilly Radar talk, made some interesting observations about the current state of application development:

• Internet, not the PC is the platform
• Apps are built on top of open source, but not themselves open source
• Doc Searl's "DIY-IT" is a key to success
• Sevices, not packaed applications
• Source code + comilation != application
• Exploring how to becone platform players via Web services APIs

The take-aways:

• FOSS doesn't guarantee freedom when applications depend on network effects and data lock-in more than on software secrets.
• Invite your used to build your services and your data, not just your code.
• if you're committed to openness, set bold standards for user control of data.
• Architect for participation via a web of smaall advantages.

11:32 AM | Comments () | Recommend This | Print This

OSCON 2004: Paul Graham on Great Hackers

What follows are some thoughts from Paul Graham's talk last night.

Variation in wealth is a sign of variation in productivity. Low-tech societies don't have variations in connectivity.

We need to understand especially productive people. How do you recognize them? How do you become one? How do you get them to come to work for you?

Like all craftsmen, hackers like good tools. They refuse to work on projects that use the "wrong" tools.

When you decide what infrastructure to use on a project, you're not just making a technical decision. You're making a social decision. The quality of the people who work on your project will depend on the tools you choose.

Programming languages are mediums of expressions. They are more than standards.

Good hackers insist on control. That's why they prefer open source.

After software, the most important tool to a hacker is the office. Offices are places to think in. Making hackers work in a noisy environment is like having a paint factory where the air is filled with grit and dirt.

What tools do hackers choose when they can choose freely? Open source operating systems and languages like Perl and Python.

People like to work with people with high standards. Buts its not enough just to be exacting. If you're not a hacker, you can't tell who the good hackers are. To drive design, the manager must be the most demanding user of the company's product.

Solving a lot of nasty little problems is not nearly as interesting as solving a few big problems. Solving nasty little problems doesn't teach you anything. Big problems have patterns. Solving nasty little problems makes you stupid.

Great hackers like other great hackers. Good hackers clump.

Having great hackers is not in itself enough to ensure success. A company that can attract great hackers has a definite advantage.

How do you know when you meet a great hacker? It seems the only way to tell is to work with them. This is why Universities are at the center of high tech areas. They serve as intellectual dating services.

How do you become a great hacker? The key to being a good hacker may be to work on things they like. To do something well, you have to love it. If you're worried that your current job is rotting your brain, it probably is. Curiosity is listed as the number one property of great hackers given by many people. The ability to concentrate is also frequently mentioned. If may not be possible to cultivate these qualities, but it is possible to repress them.

11:32 AM | Comments () | Recommend This | Print This

July 27, 2004

Exploding the Enterprise Panel Audio Available

Doug Kaye has posted the audio for the Exploding the Enterprise panel from Supernova 2004 on IT Conversations that I moderated.

3:21 PM | Comments () | Recommend This | Print This

Dave Weinberger on Blogging's Rubicon

Dave Weinberger is one of the convention bloggers at the DNC this week. He believes that blogging the convention has changed how blogging is viewed in some fundamental ways. He writes:

The credentialed bloggers are sitting in the section of the bleachers designated "Blogger Boulevard." Want to know exactly where it is? Easy: It's on the other side of the Rubicon.

This event marks the day that blogging became something else. Exactly what isn't clear yet, and the culture clash is resulting in public functions that, because there is no single culture of blogging, are Dostoyevskian in their awkwardness.

From Boston.com: Blogging crosses over
Referenced Tue Jul 27 2004 16:07:10 GMT-0600

Dave goes on to talk about the blogger breakfast that the DNC threw for them. Fascinating.

3:09 PM | Comments () | Recommend This | Print This

Arrived at OSCON2004

I arrived in Portland a few hours ago for OSCON 2004. Today is all tutorials and I'm not signed up for any of them, but tonight is Larry Wall's famous State of the Onion talk and a talk by Paul Graham on his new book Hackers and Painters: Big Ideas from the Computer Age. I've got considerable respect for Paul's ideas and his accomplishments. I bought the book earlier this afternoon and I'm looking forward to the talk. Here's a link to his IT Conversation talk with Doug Kaye.

3:02 PM | Comments () | Recommend This | Print This

City Manager and Police Chief Blogs

Dave Fletcher points to the weblogs of Eden Prarie, MN's chief of police and city manager. Reading through the first few entries, I feel like these folks are friendly and ready to help me. Thhe photos of various events and city employees are great. I think this is a fantastic use of weblogging in the public arena.

2:55 PM | Comments () | Recommend This | Print This

Jon Udell Moves to Bloglines

Jon Udell has switched to Bloglines for his news reading. His note on the switch talks about several of the features he likes, several of the features he wants, and concludes with some interesting speculation on what Google is building and a reference to BEA's alchemy

8:37 AM | Comments () | Recommend This | Print This

July 26, 2004

Google Affected By Today's Virus Storm

I'm getting slammed today by virus-containing emails sent by the latest version of MyDoom. Most of the emails pretend to be from the system administrator of the recipients email service saying, in essence, "you've been sending out virus containing emails. Open tthe attachment to fix your computer." Apparently, I'm not the only one. Google was affected as well. Apparently the latest version uses Google to search for valid email addresses within domains found on infected computers.

11:40 AM | Comments () | Recommend This | Print This

July 22, 2004

Convention Blogging

Those of us who spend our days in blogs have been seeing this develop for weeks, but many people are not aware that there's going to be some interesting happenings at next week's Democratic National Convention in Boston. Here's a few:

• Dave Winer and others have been credentialed to attend the convention and cover it in their blogs.
• Technorati will be providing real-time analysis of the political blogosphere at next week's Democratic National Convention in partnership with CNN.
• Britt Blaser, Donovan Janus and Rhesa Rozendaal have come uo with a convinient system for bloggers to post photos of the convention.

I'm interested to see what the coverage from the bloggers is like and how it gets used, aggregated, and reported.

4:20 PM | Comments () | Recommend This | Print This

July 21, 2004

Chris Warner and Amber Alerts

Chris Warner, who I've written about before with respect to Earths911.org has launched a new project for doing Amber Alerts. He launched it at this month's meeting of the National Governor Association. I saw it (and Chris) on ABC's Good Morning America yesterday. This is a very important and interesting project. The potential goes well beyond Amber Alerts to all kinds of alerting from government. I put together a presentation on the use of event notificationin Homeland Security over a year ago. Earth911.org's stock in trade has been connecting local government providers of information with local people who need that information using a nationally-based information platform. This new project capitalizes on that perfectly.

1:31 PM | Comments () | Recommend This | Print This

TA Needed for CS462

I'm looking for a TA for CS462 for the Fall. I prefer students who've had the class for obvious reasons, but familiarity with Linux is key. The class will be run on a UML environment and I need someone to sort that out starting sooner rather than later, so if you're available this Summer for a few weeks, that would be helpful too.

9:18 AM | Comments () | Recommend This | Print This

RSS Growing Pains

Both Nathan Stocks and Wade Billings took time to send me a link to Chad Dickerson's article on RSS growing pains. Chad's point: RSS as currently architected doesn't scale. RSS scalability has a few things going for it, most notably it's built on to of the Web which has shown to be hugely scalable. The problem has more to do with client implementations that don't always issue conditional GET requests or respect the resulting 304 Not Modified responses. Also, as Chad points out, many clients are set by default to check for changes at the top of the hour rather than spreading the requests out. The more basic problem is that RSS is built on a polling paradigm where clients continually poll the server to get changes. None of these problems are unsolvable and frankly, its nice to have scalability problems. It's a sign of success.

7:27 AM | Comments () | Recommend This | Print This

Responding to Utah's eVoting RFP

I joined with some other people to draft a response (PDF) to Utah's eVoting RFP. There are a number of problems with the RFP, including the fact that it does not require a voter verifiable paper audit trail and does not allow sufficient time for the State and others to adequately review vendor proposals. The response was the subject of a short article in the Salt Lake Tribune this morning. If after reading the response, you feel like you agree and support it, send me an email with your full name, address and other contact information and I'll see that you're included in a letter to the committee from others who agree with the response.

6:51 AM | Comments () | Recommend This | Print This

July 19, 2004

First eVoting Lawsuit by a Candidate

Linda Soubirous lost the March 2004 race for a seat on the Riverside County Board of Supervisors by 45 votes. She requested a recount and asked for 44 pieces of information pertaining to the recount, including the audit logs, the redundant memory stored in the machines, the results of "logic and accuracy" tests, and the chain-of-custody records for the system components. The Riverside Registrar of Voters, Mischelle Townsend refused to give the additional information saying it was "irrelevant." Now Soubirous is suing Townsend. I hope the Utah Voting Equipment Selection Committee is paying attention. I think that this suit is only the first of many if voting equipment is not selected in accordance with the highest possible standards (something the current RFP falls far short of).

1:35 PM | Comments () | Recommend This | Print This

Quantum Cryptography Network Debuts

Until now, quantum cryptography has been a strictly point-to-point affair. Technology Research News announced today that a 6-node network connecting Harvard, Boston University, and BBN has been successfully deployed. Quantum cryptography uses the quantum state of photons to transmit one-time pads to a correspondent. Reading the state changes it. The benefit is that quantum-based cryptographic methods can be shown to be strongly secure without having to base those arguments on the intractability of certain mathematical problems as standard cryptography does. (read a tutorial).

1:27 PM | Comments () | Recommend This | Print This

July 17, 2004

A Few New Mac Tools

I found a couple of new tools for my Mac this week that I like pretty well: iClip and BluePhoneMenu

The first is a handy clipboard tool called iClip. I frequently copy multiple items that I want to paste back into another document. With a multi-bin clipboard, you can do all of your cutting at once and then paste them in in another document without flipping back and forth. I was using Copy-Paste-X, but it has some funny behavior in menus that I didn't like. iClip is also a little more visual. I like how it works. (Cost: $19.95)

The second tool is only useful if you have a Bluetooth phone. After reading my post from yesterday about my new T637 Colin Kelly (from the award winning Connect Magazine) sent me a tip about BluePhoneMenu. BluePhoneMenu puts an icon in the menu bar that reads information from your Bluetooth phone. When a call comes in on my phone, the app flashes the caller ID info in the screen and pauses iTunes automatically. It also creates a log of calls on my machine and lets me select from the log to return a call or send an SMS message. Very handy.

As I was thinking about how handy having BLuetooth connectivity to my phone is, I was wondering why there are no land-line phones with Bluetooth. Afterall, I'd like use Address Book or BluePhoneMenu to drive the phone in my home office or the Cisco VoIP phone at BYU. Just adding Bluetooth to these devices would instantly give them a standards compliant, wireless interface to my computer. That's pretty simple and cheap convergence. There seem to be a few products in that space, but its hard to tell exactly how they'd work.

12:56 PM | Comments () | Recommend This | Print This

July 16, 2004

Building Models

I was writing my monthly column for Connect Magazine today and as part of that effort reread several times my recent post on Alan Kaye. As I thought about his statements about the real power of computing being their ability to do simulations, I started thinking about Stephen Wolfram's A New Kind of Science and his theory that real world processes are, essentially, just computations. One of the things I need to add to my list of things programming teaches you is "modeling." People who learn to program learn to build models.

2:20 PM | Comments () | Recommend This | Print This

Rock Regan Resigns

Rock Regan has resigned as Connecticut CIO effective Aug 1 due to "an administration change." Rock's a great guy and was a good CIO. He was CT's CIO for seven years. I wish him well.

2:02 PM | Comments () | Recommend This | Print This

Cert Says Stop Using IE

A recent CERT advisory has recommended that users stop using IE as their browser.

Use a different web browser

There are a number of significant vulnerabilities in technologies relating to the IE domain/zone security model, the DHTML object model, MIME type determination, and ActiveX. It is possible to reduce exposure to these vulnerabilities by using a different web browser, especially when browsing untrusted sites. Such a decision may, however, reduce the functionality of sites that require IE-specific features such as DHTML, VBScript, and ActiveX. Note that using a different web browser will not remove IE from a Windows system, and other programs may invoke IE, the WebBrowser ActiveX control, or the HTML rendering engine (MSHTML).

From US-CERT Vulnerability Note VU#713878
Referenced Fri Jul 16 2004 09:01:05 GMT-0600

If you're a Mac user, of course, you've already done that since IE on the Mac is far behind Safari or Firefox. If you're a Windows user, you can try Mozilla if you want a full featured browser with built-in email and address book, or Firefox for a fast browser-only solution.

Oliver Rist at InfoWorld has a column on the pros and cons or switching at the enterprise level. The summary is that you have to give up ActiveX, but you gain in security without giving up anything in general usability or user productivity.

If you haven't stopped using IE in your enterprise, now might be a very good time to give it serious consideration.

9:04 AM | Comments () | Recommend This | Print This

iSync Problems

I recently got a new phone, a Sony Ericsson T637. The phone is small, has some good features, including a crappy camera and, most importantly, Bluetooth. It had been a few months since I'd had a phone with Bluetooth, so I hadn't used iSync to sync my phone and Address Book contacts for a while. I added the phone as an iSync device and clicked "Sync Now" but I couldn't get syncing to work. It had worked flawlessly on my old T68i. Now it would sometimes sync "All Contacts" and sometimes not and it would never sync a group. I've got over 2000 cards in my Address Book, so All Contacts was not a good way to go--the phone ran out of memory and I ended up with too many contacts in the phone for the primitive navigation features it provides. So I've been trying various things for the last few days, but this morning I solved it. The solution was to completely reinstall iSync. Here's how I did it:

• Delete "iSync*" from /Library/Receipts/
• Delete the iSync application from /Applications/
• Delete "com.apple.isync.*" from ~Library/Preferences/
• Delete /Library/Application Support/SyncService/
• Delete ~/Library/Application Support/SyncService/
• Delete /Library/Receipts/iSync_Palm.pkg (I'm not using this anymore)
• Delete Ê/Library/Application Support/Palm HotSync/Conduits/Apple Ê

Then I downloaded iSync 1.4 again from Apple and reinstalled it. After adding my phone as a device in iSync, everything worked like a charm.

8:44 AM | Comments () | Recommend This | Print This

July 15, 2004

Blogging as Open Source Intelligence

This afternoon on All Things Considered, I heard a piece on Open Source Intelligence, a project at New Mexico State University. The idea is to use publicly available information and student "analysts" to create meaningful intelligence information. I got to thinking as I heard the piece that in many ways, that's the power of blogging. People use publicly available information and critical thinking to produce useful aggregations of information from sources that interest them. Some of that is better quality and more useful than others, but the idea is there.

8:50 PM | Comments () | Recommend This | Print This

Google Says the New York Times Is Irrelevant

No, Eric Schmidt hasn't been dissing the New York Times, but his search engine has. According to a Wired article, the New York Times, and other mainstream media outlets, never ranks very high in searches about various timely news topics. If you've been following along for the last several years, you know why: the pay curtain.

Of course, like many things about the business operations of a traditional publisher that has ventured online, the reasons are simple but the solutions complicated. The New York Times requires that its users register, which makes it difficult for search engines to spider its content. Perhaps an even more impenetrable barrier is the Times' paid archive. Because it stows material more than a week old behind an archive wall, you have to cough up $3 per article. Since few are willing to pay for content they can get free elsewhere, search engines, which often base results on relevancy (read: popularity), will continue to dis the Times -- as well as other media sites that make you register or pay for old news (The Washington Post, The Wall Street Journal).

From Wired News: Searching for The New York Times
Referenced Thu Jul 15 2004 14:29:50 GMT-0600

Of course, the Times could fix this. They have a huge archive. Opening it up would result in the creation of a site that is very relevant, heavily linked, and widely read. They'd get influence in spades. The pay curtain accounts for only 2 to 3% of the Times' revenue, so it wouldn't even be that expensive. Had to let go of that revenue when its in the bag, however.

At present, the offline sides of newspapers make considerably more than the online versions. For example, the Times rakes in $11 per reader in the online world and $900 per reader offline. Guess what they pay attention to. The problem is that the $900 per reader may not be sustainable as more and more readers turn to online delivery of news. That presents a business problem of incredible complexity to the Times and other print publications.

I doubt that print media is going to take this lying down. Watch for them to try to redefine how search engines work in order to support their distribution model in much the same way that the RIAA is trying to use DMCA to redefine copyright to support their outdated distribution model for music.

2:38 PM | Comments () | Recommend This | Print This

July 14, 2004

Utah Driving Records Available Online (Almost!)

Utah driving records are now available over the Internet. Unfortunately, its not free ($7.25) and its only available for your own record. Driver records accessed online contain the following information:

• Driver's name and date of birth
• License number, type, and status
• Original license issue and expiration dates
• Commercial status and license endorsements
• Moving citations (citations issued for three years for non-alcohol and 10 years for alcohol-related violations)
• Arrests (all DUI arrests received in the past 10 years)
• Department actions

When I tried it, the system happily took my credit card data and then said:

An error has occurred. If you continue to have problems, contact customer support by going to: http://www.utah.gov/contact.htmlThere are no error details available.

That's special, isn't it. Of course if Utah Interactive takes $7.25 from me for nothing it won't be the largest amount of money I've given them for no good reason by a long shot.

11:24 AM | Comments () | Recommend This | Print This

Tracking Mexico's Attorney General and Japanese School Children

Mexico's Attorney General now has a non-removable microchip implanted in his arm to give him access to the countries crime database and track his movements. The device will be deactivated after he leaves office. I guess if someone kidnaps him now they'll be forced to cut off his arm to keep from being tracked.

In a similar move, although for different reasons, Japanese authorities in Osaka have decided to use RFID devices to tag school children. In this case, the devices will just be attached to uniforms and backpacks. I have to admit, I'd be willing to put them on my kids.

11:14 AM | Comments () | Recommend This | Print This

The Value Proposition of Software Assurance

eWeek has a story relating that volume licensees are pushing back on Microsoft's Software Assurance program. When I was CIO for Utah, Microsoft pitched us on this and we refused to bite. Here was how I saw the value proposition: "We need money from you every year to keep our stock price up. We've been really clever at squeezing money out of you every year. So, let's save both of us a lot of hassle and you just agree to send a bushel basket of money to Redmond once a year." When sales tactics didn't float the deal, they accused us of not being in compliance on licenses. The choices were (a) prove you're in compliance or (b) just sign up for Software Assurance. We chose (a) and because we didn't have a good asset tracking system spent a lot of time and money conducting an audit which eventually showed that we actually had purchased more licenses than we needed. Funny, Microsoft didn't offer to buy them back. Go figure. The moral of the story: make sure your asset tracking system is up to date.

10:58 AM | Comments () | Recommend This | Print This

July 13, 2004

How to Start a Blog

I'm sure there's a million of these on the net, but I get asked occasionally by friends how to start a blog, so here's a collection of tips that I wrote to a friend recently. Now they're written down somewhere...

8:35 PM | Comments () | Recommend This | Print This

VisaProcess, Meta-Mail, and Virtual Networks of Demand

On a note related to the article on Alan Kaye I just posted, I just was reading Esther Dyson's abstract for this month's Release 1.0 on what the spreadsheet did for data, freeing users from models built on the mainframe and why we don't have a similar tool for processes. She says:

First story: My inbox is overflowing. I have 3158 messages in it, dating back to the last general cleanup, January 2004. I also have a folder called Memorial Day, which contains 1825 messages dating back to spring of 2002 and before: These are all the messages I was planning to handle over the 2002 Memorial Day weekend, but never got around to. I know that I can find them, even with Eudora's relatively slow search. But I want to know more: Which ones of the 3158 new ones should I be paying attention to and looking for?

Second story: A couple of weeks ago, analysts following Omnicom Group noted that the company plans to spend an extra $50 to 60 million in audit fees and internal costs (mostly IT, we assume) to comply with the new Sarbanes-Oxley requirements. Presumably it has all the data, but now it needs to make the processes explicit: Who's in the chain of command? Who made the decision to pay the bill? It wasn't made by the programmer who wrote the code.

Two stories, one theme: Getting control of business processes, not business data.

Indeed, data is relatively easy, and we have good tools for it: the calculator, the spreadsheet, and the giant financial number-crunching application. The spreadsheet gave users a tool not just to calculate, but to build complex models and, in fact, to do many things that previously could be done only by IT high priests. Better yet, the spreadsheet allowed them to build models that were intelligible to normal people. So-called power users could build the models, while other users could reuse or modify them, plugging in their own data and coefficients. Complementary graphing and other tools made the data more visible and meaningful to ordinary people who could not pick trends out of a sea of numbers. We also have the database, which acts as a back-end to those corporate applications and to the spreadsheets, allowing for easier sharing of data across applications and even among enterprises.

The first successful spreadsheet was called VisiCalc; where is VisiProcess?

From Business Intelligence & Market Research with Technology Analysis- Release 1.0 Publication
Referenced Tue Jul 13 2004 11:47:52 GMT-0600

Esther also uses the term "meta-mail" to describe this. I've used a different term for a similar idea in the past: virtual networks of demand, a term I first heard from Duy Beck. Here's the idea:

Anytime you get an organization of more than a few people, you start hiring people with particular functional specialties to perform specific tasks. Therefore, getting anything accomplished, requires that you have a workflow (formal or informal) for getting things from one person to another in the right order, at the right time, etc.

Another way to think of this is as every person representing a little bit of production capacity with their own supply chain and demand chain. All of these internal supply and demand chains represent a "virtual network of demand." The goal of an organization is to find ways to efficiently and effectively service this network and keep it flowing.

From an IT perspective, when we install CRM systems, ERP systems, employee portals, workflow systems, personal computers, office suites, and the like, we're trying to service and automate this demand network. The problem is that we can't, yet, approach it from the standpoint of viewing each employee as a custom unit that has specific needs because of their role, their style of work, the way they learn, the way that they're most comfortable communicating, etc. We more or less give everyone a standard set of tools and require them to do their own customization. We just build a machine and expect people to be cogs in it, instead of viewing them as a big distributed P2P network. I think we'll see a trend in the future toward more and more fine grained approaches to this problem.

From Phil Windley | Virtual Networks of Demand
Referenced Tue Jul 13 2004 11:49:43 GMT-0600

Do things like BPEL get us closer to this? I'm not sure. Its not that BPEL isn't a good thing, but that its just one piece and incremental at that. What made the spreadsheet so revolutionary is that it let non-programmers build sophisticated models by putting programming into a rigid enough box (no pun intended) that many of the gotchas we experience in programming were no longer relevant. I'm not sure BPEL (or other process oriented languages), even with a great visual programming front-end, represent this same kind of fundamental leap forward.

11:59 AM | Comments () | Recommend This | Print This

Alan Kaye and Modeling

Fortune has an article on Alan Kaye. Alan decries the current uses of computing and says:

"We're running on fumes technologically today," he says. "The sad truth is that 20 years or so of commercialization have almost completely missed the point of what personal computing is about."

From Fortune.com - Fast Forward - A PC Pioneer Decries the State of Computing
Referenced Tue Jul 13 2004 11:28:46 GMT-0600

His chief complaint with regard to business use of computing is that its more or less simply used as a replacement for pen and paper. What he'd like to see more of is modeling and simulating. He doesn't mention spreadsheets, so I'm not sure if he sees them in the same light or just the way their commonly used. Whenever I contemplate a new business one of the things I do is create a free cash flow model of the business and run lots of scenarios. I don't think I'm alone in that and that's certainly modeling.

I am, however, shocked by how difficult it is to show those models to other people. When I do this, most people want a static snapshot printed out and stuck in a book. That hardly captures the dynamic nature of the model.

Alan's latest project is Squeak. Alan demonstrated Squeak at ETCon 2003. Here's what I said at the time:

The keynote for this morning is Alan Kay and is entitled "Daddy, are we there yet?" Alan is the inventor of SmallTalk, among other things and he has Utah ties, getting his PhD from the Univ. of Utah in 1969. His primary complaint is that the last 20 years have been pretty darn boring because we're spending our time making better buggy whips in the form of better spreadsheets, better ways to write memos, etc. Alan quotes from a paper written in 1963 called "Man-Computer Symbiosis:" In not too many years, human being and computing machined will be coupled together very tightly and the resulting partnership will think as no human brain has ever thought before. he claims that this hasn't happened in large part yet, although I'd argue that networked computing has allowed some of us have made some progress in that direction.

To prove his point, Alan is going through four or five examples (all with live demos and video which is pretty cool) of really neat, revolutionary computer science breakthrough including Ivan Sutherland's Sketchpad and McCarthy's LISP 1.5 implementation for the PDP-1. He is now showing what is likely the first collaborative software system: Engelbart's NLS. Built in 1968, it had live full motion video conferencing and concurrent document editing with shared cursors.

Alan shows something called "end user computer literacy" which is a graphical LOGO-like environment where users draw things like cars and steering wheels and then connect them together using scripts they build from property panels. It doesn't sound dearly as impressive as it was to see. The goal is to get kids building programs and using it to do experiments. One of the examples is using the system to explore the notion of a gravitational acceleration constant.

The base system is called Squeak. Squeak is the programming language and the OS. Its a grand total of 2.8M of binaries. One interesting thing he showed was that his presentation was actually a sequence of desktops so that each slide was a fully functional view. This made for some pretty cool demos.

From Phil Windley | Done at ETCon
Referenced Tue Jul 13 2004 11:38:33 GMT-0600

11:39 AM | Comments () | Recommend This | Print This

July 12, 2004

Utah's Electonic Voting RFP

An editorial in today's Deseret News urges the state to proceed slowly with a move to electronic voting. The Voting Equipment Selection Committee (VESC) released an RFP last week. You can read the press release (PDF), the executive summary (PDF) or the full RFP (PDF) for yourself. I loved this quote from the press release:

ãMembers of the Voting Equipment Selection Committee have worked diligently to develop an RFP that addresses both the needs of Utah voters and the requirements of federal and state law,ä said Lt. Gov. Gayle McKeachnie. ãBy leaving the door open for numerous systems including optical scan and electronic voting equipment, I am confident the state will select a system that Utahns will feel comfortable using.ä

The Lt. Governor got it exactly right. VESC has left lots of doors open, seemingly unwilling to actually take a stand on almost anything. Usually you write an RFP to make sure the State gets what it needs. This RFP seems crafted to invite vendors to tell the State what it needs.

I'm particularly concerned about the lack of a requirement for voter verifiable audit capability in electronic solutions. The RFP makes a nod in the direction of security and says that's the second most important selection criteria (after cost), but saying your concerned about security in general terms without being quite specific about what you require leaves the door open for all kinds of rationalization. In fact, if you do the math, security counts about 10% while cost counts for 30% in the selection. That seems upside down to me.

One need look no further than the constant plague of email viruses and security warnings to understand that computers suffer from a whole host security problems. The problem is that security is not something you can guarantee with a few code reviews or a carefully chosen panel of experts (as the RFP suggests). In fact, security is so hard to achieve that most Computer Scientists who have looked at the problem have concluded that the only way to prevent problems is to provide a method for auditing the results, provide for independent recounts and to allow each voter to check the results of the machine. This is called a voter verifiable paper audit trail (VVPAT).

California Sec. of State Kevin Shelley is concerned enough about security issues in electronic voting machines that he's banned (PDF) the use of machines that don't contain a voter verifiable paper audit trail. I don't understand why Utah's VESC has been unwilling to take a similar stand. They've never said as far as I can tell. In fact, you'll have a hard time finding any information that will help you understand their rationale for their decisions or what they think is important. It seems as if their prime directive has been to keep everyone happy and get an RFP out:

ãThe Voting Equipment Selection Committee has been very deliberative in drafting this RFP,ä said committee chair and state Chief Information Officer Val Oveson. ãWe have been very collaborative in addressing all issues related to the purchase of a new statewide voting solution and are excited to see the RFPâs timely release.ä

Unfortunately, when it comes to computer security and especially in an area as important as ensuring the integrity of the voting process, you need the help of computer professionals. This isn't an area where consensus alone will necessarily lead to a better result.

Now that the RFP is out, our main hope is that VESC will be very careful in applying their security criteria. As they proceed through the evaluation process, the public will have a chance to comment on the various proposals. I urge you to get involved and let VESC know that you only support the selection of electronic solutions that have a voter verifiable paper audit trail by signing this online petition if you live in Utah.

8:54 AM | Comments () | Recommend This | Print This

July 9, 2004

Bluetooth Enabled Luggage Tags

This is just a concept, mind you, but its a good one. In response to an industrial design contest at IDFuel, Nathan Lynch and Lea Miller had proposed a Bluetooth enabled luggage tag that lights up when the cell phone its paired with is nearby and sends the phone a message. The tag also contains contact information for the owner of the luggage. Essentially, this is presence for physical objects and there are all sorts of applications that spring to mind.

1:33 PM | Comments () | Recommend This | Print This

The Fools are Fools

The Motley Fool has an article on the INDUCE act and in reference to people claiming that the new bill would kill the iPOD, they say "Don't you believe it." Here's the quote:

Don't you believe it. While I'm normally not a fan of legislative remedies to technical problems -- and I've got the public ranting to prove it -- this bill doesn't seem all that unreasonable to me. Of course, I don't steal -- I mean, file-share.

From Fool.com: Will Congress Kill the iPod? [Motley Fool Take] July 8, 2004
Referenced Fri Jul 09 2004 09:34:36 GMT-0400

This whole argument is a red herring. The question isn't whether Apple and other manufacturers will stop making MP3 players. They won't. The question is, if this bill had been in force 5 years ago, would any company have built an MP3 player? I think that reasonable people can conclude that businesses would have been more cautious and we wouldn't have the market we have now.

What opponents of the bill have argued all along is that it stifles innovation. Given that, I think it ironic that Sen. Hatch is sponsoring legislation that disadvantages the kinds of business that Utah has in spades (small innovative firms) in favor of entrenched businesses (which Utah has few).

9:43 AM | Comments () | Recommend This | Print This

Learning PHP

Dan Olsen links to some Web-based tutorials, documentation, and books he's found useful in learning PHP.

Also, Jeremy Zawodny is arguing against abstraction layers in PHP, something that seems right to me. When I first read the title, I thought of Hibernate and other object-to-relational mapping tools. I'd argue that those abstraction layers are heaven sent. I don't think Jeremy would disagree since ultimately his argument comes down to the fact that PHP data abstraction layers don't really offer meaningful or useful abstractions--something that's not true of object-to-relational mapping layers.

9:12 AM | Comments () | Recommend This | Print This

Doc Gives Lessons in PR Writing

Doc Searls has written an excellent rant about PR email releases over at IT Garage, including tips on how to write this kind of tripe. Since I started writing for InfoWorld a year ago or so, I've started to get these too. They are just short of SPAM. The example Doc quotes is not unusual or rare. There are a lot of companies who pay good money to PR firms to generate these and send them out to people who are subsequently annoyed by them. Go figure.

9:09 AM | Comments () | Recommend This | Print This

Take Control of Your Airport Network

If you're setting up a wireless network for your Macs, Glenn Fleishman's new eBook may be what you're looking for. You can't go wrong with Glenn's expertise and the book's inexpensive price.

Also, Apple has put up an article about creating an extended Wi-Fi network using AirPort Extreme and Express base stations. I got a couple of the new Express units on order and am anxious for them to make their way to my inbox.

8:58 AM | Comments () | Recommend This | Print This

Tom Malone Speaks on the Future of Work

Doug has posted Tom Malone's Talk about the future of work at IT Conversations along with a transcript. If you can't read the book, then listen to or read the talk. I think this is important perspective on technology's impact on the business.

8:24 AM | Comments () | Recommend This | Print This

July 8, 2004

The Pentagon's Messy Backoffice

The Pentagon has backoffice problems. Over the last decade, the Pentagon has spent $19B dollars in creating numerous systems to help manage accounting and logistics at the Dept. of Defense and things are still broken.

Some outside analysts see the inefficiency as an unfortunate but necessary consequence of the Pentagon's enormous commitments and largely successful track record. But others think the Defense Department could handle its operations a whole lot better.

"If you ran your business this way, you'd be in jail," said Christopher Hellman, an analyst with the Center for Arms Control and Non-Proliferation.

From Wired News: Data Nightmare at Pentagon
Referenced Thu Jul 08 2004 16:26:04 GMT-0400

The problem is that the Pentagon isn't a business. It's a government agency. I found out, some would say too late, that there's a difference. The difference is in motivation. Government agencies are mission oriented, not profit oriented. The Pentagon does not see its job as being efficient. That's not to say that they don't think its important to be efficient, but there's no natural driver in that direction.

There's no doubt that where the military sees information systems as critical to their mission, that they can develop and deliver effective systems to the task. They do it time and time again. The irony, for me, is that these backoffice systems may be the key to fighting in the future. As we move our defense priorities to defending vulnerabilities, the inventory, supply and logistics expertise and efficiency may be as important to our defense as more offensive capabilities. I predict that when they become mission critical, these systems will get cleaned up in a hurry.

5:08 PM | Comments () | Recommend This | Print This

July 7, 2004

Computer Science Isn't Programming

Jim Morris, who is a professor of computer science and dean of Carnegie Mellon University's West Coast campus has an article on Computer Science education in the Pittsburgh Post-Gazette. He says:

the vocational nature of computer science reduces its appeal to many students. Contrast what computer careers seem to offer with the promise of the traditional sciences that offer intellectual grandeur and the opportunity for a rewarding career that helps humanity. Since 1990, the number of undergraduate degrees awarded in the biological sciences has increased 70 percent. During the same period, the number of computer science degrees awarded each year has dropped by 10 percent. Today, we are producing about 25,000 computer science bachelor's degrees annually. The life sciences produce six times as many.

The current approaches to computer science education fail to teach the science of computing. As a result, they fail to inspire the very best and brightest young minds to enter the field.

From Programming doesn't begin to define computer science
Referenced Wed Jul 07 2004 17:52:43 GMT-0400

I've got at least one corroborating data point: when the boom was happening, I had kids come to my office all the time, usually after their second year, when they get the living tar beat out of them with several large programming projects, and say something like "I don't like programming. Should I stay in CS?" I've had various feelings about that.

I usually told them that I knew lots of people who had degrees in CS but didn't program for a living. Some were system administrators, network engineers, lawyers, or doctors and lots were salespeople.

As Morris says, there are lots of interesting challenges in CS that go well beyond programming, but its difficult to appreciate them without at least an understanding of programming. I usually advised students that even if they didn't end up programming for a living, learning to program would give them a special perspective. Here are a few things I think they might get out of programming:

• People who program have a unique understanding of the relationship between static representations and dynamic processes (I have a theory that music composition students get a similar understanding).
• Participating in programming projects teaches you why its hard to get software right. You understand the non-linearity and brittleness of software.
• Programming as part of a CS degree should also teach students (although I'm afraid sometimes we fail) why learning to write small programs doesn't qualify you to engineer large software projects. We do a worse job at teaching how large software projects can be successful.

What other things do you think learning to program teaches a CS major even if the CS major isn't going to program for a living? I'd like your perspective.

6:11 PM | Comments () | Recommend This | Print This

Gartner's Barking Up the Wrong Tree

Eric Nolin links to the Gartner reporton the iPOD and other firewire/USB storage devices. Eric's point is that this is one of the issues driving Web (or NET) 2.0. I agree. I think, however, that CIOs need more help than just saying "ban firewire/USB storage devices." This is the standard perimeter approach to protecting corporate data.

The problem with that approach is that its not feasible and getting less so all the time. I've got at least five devices in my laptop bag that could be used, potentially, to carry data out of a secure perimeter. Are you really going to tell people they can't bring cameras to work? What about cell phones? Watches? Heck, burning CDs is relatively easy with the standard corporate desktop. Some might think DRM is the answer, but its not. DRM is high overhead and requires careful planning of who has access to what. That's fine for high value data objects, but its not a general solution.

What businesses need is a way to audit user actions. I ought to be able to ask what the history is for any document and see not only the details about how this copy has been modified, but when and to who it has been emailed and any copies that were made, by who and to where. Processing audit logs scales linearly and allows you to focus attention on problem areas after the fact. Simple, workable, and solves most of the problem.

12:02 PM | Comments () | Recommend This | Print This

July 6, 2004

Rounding Up Web Services Intermediaries

As vendors fine-tune their products, add features, and improve usability, WSI options are getting better. Over the last few months, several new versions of WSI products have popped up: Actional Looking Glass 5.0, AmberPoint SLM (Service Level Manager) 2004, Grand Central BSN (Business Services Network) 4.0, and Infravio Ensemble 4.1 (the component I tested is now known as X-Broker).

Each of these products differs significantly in their approach to intermediating Web services, but a cursory review of their feature sets doesn't readily reveal that difference. The significant contrast is in the metaphors they use for management and the presentation of information; each WSI emphasizes different aspects of the intermediary game. Grand Central 4.0 is a hosted service, whereas Actional's approach zeroes in on service monitoring; Infravio uses contracts to define relationships; and AmberPoint SLM manages performance with service-level objectives.

From InfoWorld: Rounding up Web services intermediaries: July 02, 2004: By Phillip J. Windley
Referenced Tue Jul 06 2004 21:21:07 GMT-0400

Doing a 4-way head-to-head comparison of Web services intermediaries was a bigger job than I thought it would be. The job is made tougher by the fact that none of these products are trying to solve exactly the same problem. Nevertheless, I found them all to be quite capable with each having specific strengths. I continue to recommend Grand Central's Business Services Network to people wanting to explore the capabilities of an intermediary because its so easy to get set up and developer accounts are free.

The piece also included a sidebar on BPEL:

Because BPEL is XML based, it's not much to look at. Programming language designers call the features used to make a language readable and pretty "syntactic sugar;" BPEL is syntactic arsenic. There seems to be no compelling reason to base BPEL on XML except that XML is the syntax of the Web.

BPEL's constructs will be familiar to anyone who has programmed in a language that allows parallel execution flow and asynchronous invocation -- good programmers will catch on pretty quickly. And because BPEL is a standard supported by large vendors -- it was defined by BEA Systems, IBM, Microsoft, and a handful of other vendors and proposed to OASIS as a technical standard in 2003 -- support tools should be plentiful.

Consequently, most people probably won't write BPEL directly, but rather use a graphical tool that ultimately generates the BPEL. Grand Central's Business Services Network, for example, provides a convenient engine for executing BPEL. Collaxa sells a BPEL execution engine as well as a BPEL design tool.

From InfoWorld: Is BPEL the real deal?: July 02, 2004: By Phillip J. Windley
Referenced Tue Jul 06 2004 21:33:09 GMT-0400

9:34 PM | Comments () | Recommend This | Print This

July 5, 2004

ListGarden for Creating RSS Feeds

Creating RSS is usually the province of content management systems (such as blogging software) that not only creates an HTML page, but also generates an RSS feed as well for the same content. Sometimes, however, you just want the RSS. I've done that using Radio. I just create a category that generates only RSS and post to it. That's how I create my newsletter, for example. Easy enough. Dan Bricklin (of VisaCalc fame) has created a little tool called ListGarden that does the same thing, is written in Perl, and freely available. I tried it out and here's what I found.

ListGarden installs easily on OS X 10.3 or any other system that supports Perl. There's an executable version for Windows, but I didn't try that. The only thing I had to do to get it going was install the LWP::UserAgent package. The ListGarden product page includes instructions on how to do that if you're not familiar with CPAN. The program uses the browser for an interface, so after you run it, you just point your browser at port 6555 and you're in business.

One note of caution: if your machine is accessible through the Internet, you'll want to install ListGarden behind an HTTP authentication curtain since there's no built-in authentication scheme. The tool allows files, with any name, to be created on the local drive or any FTP area you've set up. If you install it wrong or leave it unprotected, you will be opening up a big security hole.

Setting up a feed is easy. There were a few small problems. I used an underscore in my name the first time and nothing happened. No error message and no feed. The second time I removed the underscore and hit the return key after I typed the name and got a different action than I expected. Hitting the "Create" button fixed that however.

You have to configure each feed with a name, description, and also give it publishing information. You can publish the resulting files to your local machine or FTP them to another machine. Both are fairly straightforward.

The tool also creates HTML from the RSS feed as an option and the HTML is templatable to create your own look. Here's my my test feed and the companioin HTML file. All done inside 15 minutes, including installing and configuring the software, creating the feed, and adding a few items.

Why would you want this? If you've already got a way to generate RSS, this program is probably superfluous. Still, its lightweight, easy to install and use, and works. I can think of several possible uses and I'm sure people will discover more:

• A lightweight blog. Its not MovableType, but its not trying to be. For small tasks, it would work well.
• Creating special purpose RSS feeds on your own Website. The Forum comments box on the right side of my blog shows RSS. I could use ListGarden in conjunction with such a tool to manage content boxes on my Website.
• Change logs. Think of all the Websites that list the version history for software. You could publish that as an RSS feed instead.
• Simple lists. Because the tool can produce HTML as well, just about any list, especially if it changes from time to time, could be stored as RSS and available from a browser as well.
• Alerting. Anytime you want people to be notified of something, this is an easy way to have those alerts show up in their feedreader.
• Generating RSS to feed into another program. We're creating tools that eat RSS in my lab. Being able to easily generate RSS is a help.

9:22 AM | Comments () | Recommend This | Print This

July 2, 2004

Scaling Programming Projects

Patrick Logan comments on my post about the lack of modern compiled languages for optimized performance by pointing to an excellent article by Dan Friedman and asks "Why not take the road less travelled?"

The Friedman article contains a story by Jonathan Sobel where he relates "winning" a class competition for writing the fastest program by using Scheme as the design language, performing correctness preserving transformations on it, and then finally translating the result into C. I've no doubt that you can do this because I've done it myself. I've taught Scheme for almost 20 years and my Ph.D. work and major research forcus for a decade was formal methods. The problem becomes one of scale.

If there were a small core that needed to be optimized, this wouldn't be a bad method. I could have some confidence that we'd end up with something better than we would by simply sitting down and letting people start writing in C. On the other hand, would you be willing to advise someone else to bet their company on this methodology for a large project, especially considering that the people on the project have no training or experience in these techniques? I can't do that in good conscience.

Like it or not, the world "accretes programs as series of patches" in the words of Paul Graham. That means that all the tools and experience are optimized for that activity. The "wins" for alternative programming styles are almost always presented in the small or at best by a team of experts. I know of no case studies that show large projects being done using more formal tools and methods by people who were, beforehand, unfamiliar with the techniques and unbelievers in its efficacy. Evolving to Patrick's better world is a desirable goal, but I'm not willing to bet someone's company on the uncertain, for me, outcome.

9:52 AM | Comments () | Recommend This | Print This

July 1, 2004

No Digital Ink for PDFs

About a month ago I bought a Tablet PC (HP T1100). Mostly I wanted to see how it worked, play with the form factor and so on. Tuesday I'm flying to DC for and NSF Review panel. I have about 15 proposals to comment on before I get there and they're in PDF format. I thought "The tablet is the perfect form factor for working on an airplane, especially since all I have to do is read and comment on the documents." How natural it would be to view the PDF document and write notes on it. Unfortunately, it can't be done as far as I can tell. This floors me. The Tablet's been out for two years and you still can't do digital ink on PDF documents. Wow!

7:56 PM | Comments () | Recommend This | Print This

Web Services Disproportionately Benefit Small Business

Esther Dyson says that small businesses have a higher surface to volume ratio than large businesses. What she means is that small businesses are much more dependent on interacting with the outside world than large business are. Does this mean that small businesses will disproportionately benefit from Web services and easier integration? I think that's just another way of looking at what's happening to large businesses as they outsource more and more of their support structure. They're becoming "smaller" and increasing their surface to volume ratio. There's a tie-in to what Tom Malone was saying at Supernova last week.

11:03 AM | Comments () | Recommend This | Print This

Marketecture

I heard the word "marketecture" used in an interesting way today. Traditionally, its been used as a synonym for slideware but my friend, who's the CEO of a software applications company, used it in connection with situations where he goes on a sales call to a large company and they'll only accept his product if it meets their preconceived notions about architectural purity. He called J2EE his marketecture since his architectural decisions are being driven, in part, by what his customer want to hear.

10:57 AM | Comments () | Recommend This | Print This

VMs as the Dominant Software Platform

Jon Udell describes how virtual machines have become the dominant software platform and some of its implications:

At that point something clicked in my head, and I proposed a software taxonomy based entirely on virtual machines -- the VB runtime, the CLR, the JVM, the Perl and Python VMs. Some of these are bound more tightly to operating systems than others, some are bound more tightly to programming languages than others, but they all share a set of common characteristics. The definition of a modern "software platform," I would say, is a VM and its associated class libraries. And a bunch of implications flow from that.

From Jon's Radio
Referenced Thu Jul 01 2004 07:33:39 GMT-0600

He goes on to describe how VM-based software platforms have enabled whole new ways of doing software testing.

I was noticing this from another angle the other day. I was having a discussion with some engineers about what languages to use in various portions of a project. There was a core piece that really needs to be optimized for performance to the greatest extent possible and other parts where cross-platform and user interface issues dominate. I began to realize that almost no recent development in programming languages helps with the former. We're probably still going to write it in C++, with all its warts, so that we can compile it. Will people still be using C and C++ to write operating systems and other core software 20 years from now? Doesn't seem right, but it is likely.

7:41 AM | Comments () | Recommend This | Print This

Computer Repairs by UPS

This Business 2.0 article reports that Toshiba and UPS have entered a deal whereby UPS will provide computer repair services for Toshiba computers under warranty. This is an interesting idea by itself, but think of the extension of this. We typically think of retail being about lots of specialty shops providing their service with the logistics infrastructure bringing their goods to them. This turns that inside out so that there's just one shop that takes to goods to the service providers.

7:30 AM | Comments () | Recommend This | Print This

Mono 1 is Released

Novell's Mono project, the open source C#/CLR runtime, has released verion 1.0.

7:11 AM | Comments () | Recommend This | Print This

Bye, Bye Mr. CIO Guy

Scoble points to one of the funniest things I've seen in a long time: Pat Helland singing Bye, Bye Mr. CIO Guy with Don Box on guitar and David Chappell on piano. There's six verses!. Here's a sample:

And the men that I admire least,
The MBAs trained in the East
Made sure their salaries were increased
The day that IT died


So Bye, Bye Mr. CIO Guy
Gonna outsource every resource till the business goes dry
And MBAs watch the beans flowing by
Singing "this will make the P-E go high!"
"this will make the P-E go high!"

7:09 AM | Comments () | Recommend This | Print This