Alexis Smirnov

2005-03-06T01:43:00.000-08:00

The words we say are usually defined

by colors of the wind

That travels in the dreams.

The sweetness of first denial,

The smell of seaweed,

Or by the strength of the desire.

By passion in the touch,

And stroke of brush

Across the empty canvas,

Or may be by the drop

That falls from ice

On the cheeks of your beloved.

Iroshka

2004-07-16T09:23:00.000-07:00

This picture was taken on our conttry house in Mt. Tramblant area. Lots of calibris there.

2004-02-11T19:17:00.000-08:00

Sam Ruby summarizes the lesson of using wiki, mailing list, weblog during the development of Atom :
XMLConf: Lesson 3: Use a mix of strategies

2004-02-08T15:49:00.000-08:00

Decentralized Meta-Data Strategies - fairly date, but still relevant catalog of distributed meta-data strategies usefule in resource location, distributed queries.

2004-02-08T12:12:00.000-08:00

Joel wants to link .NET framework into an .exe. This tool does it.

2004-01-15T05:40:00.000-08:00

Just stumbled on Grokker

Their product grokker works on top of a search engine to present search results in a graphical multi-layer fashion instead of usual long list of links.

Interesting UI innovation that moves users away from long listbox paradigm. Also interesting is the way they classify search results to divide entire result set into logical subsets.

2004-01-06T13:21:00.000-08:00

The Register : "SETI@home crew is dabbling with Sun's JXTA peer-to-peer protocols for future versions of BOINC. "
5M SETI@home users on JXTA network would be a good test!

2003-11-30T13:01:00.000-08:00

Adam Bosworth: I'm not the huge fan of REST that I think many in the ebpml.org world and elsewhere are. I love the ease and simplicity of REST, but once one assumes messaging is the correct paradigm, then there is need for an envelope to carry along the metadata used to correlate the message responses with the requests since this may no longer be a synchronous request/response. Secondly, in the world I'm describing, as I'll be making clear in future posts, a fair amount of information has to come in the request in order to enable the service to synchronize its response. This isn't possible if the request is limited to a URL which is limited by most systems to 2K bytes. Thus I think the argument that the REST folks make for all queries being GET's is impractical.

more...

2003-11-28T07:50:00.000-08:00

iPod's Dirty Secret - Neistat Brothers

Don't you LOVE when the anti-marketing machine is as slick as the marketing machine!

2003-11-12T07:53:00.000-08:00

Here's what i've learned from Tim's critique of Clay's critique of Semantic Web: It's ok to not try to boil the ocean.

It is unjust to dismiss a grand theory as falure when small-scale applications of such theory may succeed. Tim points to exposing machine-readable financial statements as one such application. I can think of other small-scale applications...

2003-11-11T11:35:00.000-08:00

Linux. The future is open. - IBM's marketing machine continues to impress...

2003-11-11T11:34:00.000-08:00

Critical Section - The Emperor's New Code - bookmarking a nice summary of latest PDC

2003-11-11T11:32:00.000-08:00

Tim: "essays on the construction, deployment and use of search technology, by which I mean primarily “full-text” search."

2003-11-11T11:31:00.000-08:00

eetee - JXTA secured groups & BitTorrent-Gnutella-OpenNap-eDonkey

eetee does to p2p what trillian does to IM

2003-11-11T11:30:00.000-08:00

Ted has joined Chandler team to work their repository and querying.

2003-11-11T11:28:00.000-08:00

Bookmarking for future read:
Clay talks about semantic web:
Shirky: The Semantic Web, Syllogism, and Worldview

Tim picks up

2003-11-11T06:42:00.000-08:00

From internal mailing list:
This is from Microsoft, and describes when to write an exception, what to put in it (the type of constructors, attributes, etc, etc) and also includes a utility to test the exceptions in your assemblies. Short and sweet. Highly recommended.

The Well-Tempered Exception

2003-11-06T18:30:00.000-08:00

Google Search: deskbar

Next stop: google searches your desktop

2003-10-31T15:44:00.000-08:00

cross-VM development

Russell talks about how JVM needs to grow past Java. Sure, that would be nice.

But, one of the things I love about CLR is the tools I get to use when coding against it. One thing I love about VJM is that it remains single most solid cross-platform runtime for managed code (I don't yet consider Mono solid enough).

Here's a question:
Has anyone tried to build anything using J# on Windows and Sun's JVM on other platforms.
It should be theoretically possible:
J# has the same syntax as Java. J# comes with some kind of converter jbimp.exe .jar to MSIL .dll

I've stumbled on this note while looking for prior art on this, but there's not much more in terms of experience reports.

2002-10-25T07:37:00.000-07:00

This weblog has been moved to http://radio.weblogs.com/0112946/

Please update your links.

2002-10-25T07:29:00.000-07:00

It's been a couple weeks since I started getting rare random crashes of Groove Workspace on my machine. According to MoonlightGroove (moderator of Groove support forums) this is one of those problems that just would not happen in their lab.

Having shipped 8 software products in my life I can relate to the pain support and development teams experience in trying to hunt down those kinds of problems. So I'm now installing patches, running Groove in debugger, sending call stacks and memory dumps to Groove Support and waiting catch that rare crash. The point is, if you care about a product - help its developers to help you.

MoonlightGroove announced yesterday that the latest patch release 2.1b fixes some of the crashes.

As an added bonus, I found activation code for Groove Professional Edition in my in-box this morning as a birthday gift from MoonlightGroove. What a wonderful surprise!

2002-10-23T06:13:00.000-07:00

Being true to the duty of "Master of Ceremony" Chris Sells posts a good collection of Web Services DevCon summaries from various weblogs.

2002-10-22T07:00:00.000-07:00

High Tech Start Up: The Complete Handbook for Creating Successful New High Tech Companies
by John L. Nesheim

This book is for anyone thinking about starting a high-tech business and is hoping to finance it. The author does a great job exposing all the complexities of finding the money and closing the financing. This book shows the problems to be watched for at every step of the way. But notice one thing - this book was written that time of common VC and high-tech lunacy. Now when first-to-market-wins rules of the game are no longer applicable. I wonder how the author would change the tone of the book if it was written in 2002.

Regardless of the times, this book shows just how hard it is to get the money and why the financing should be done only if absolutely necessary.

While this book does a great job covering all-things-money, it is not a complete handbook of creating companies. This book does not cover all of the other aspects CEOs need to worry about. If you're looking for a comprehensive handbook, check out book by Gordon Bell.

2002-10-21T12:27:00.000-07:00

Kenny Jones, Martin Harwar, and Edward Jezierski have put together a 142 page paper called "Deploying .NET Applications. Lifecycle Guide" availabe on MSDN and in PDF.

This guide provides guidance, recommendations, and the technical information you need to successfully plan for and deploy Microsoft .NET applications.

This is the most comprehencive .NET deployement work that I've seen so far.

2002-10-17T14:15:00.000-07:00

NUnit V2.0 is released

Some of the highlights of the new release include the following:
- Attribute based mechanism for specifying test and test fixtures.
- New Forms Interface that displays Tests and TestSuites in an
Explorer like fashion. The interface allows you to run individual
tests and/or suites from the forms interface.
- Automatic construction of suites based on namespaces. Just provide
an assembly and the test runner constructs a suite of all
TestFixtures in the assembly.
- Minimal amount of effort to upgrade due to inclusion of a
backwards compatibility class called TestCase and the framework also
looks for methods that begin with "test" regardless of case as it
did in the past.
- XML output from the console program.

[DOTNET-PRODUCTS@DISCUSS.DEVELOP.COM]

2002-10-17T13:14:00.000-07:00

WS DevCon: Tim Ewald: Building Web Services using System.XML

Tim is a big proponent of what he calls ?XML Way? of doing web services. This basically means using XML (as oppose to object models) as data representation for a web service. His weblog has a note explaining the concept.

Tim didn?t only talked about it, he also presented some practical solutions with example code that shows how to customize C# [WebMethod] (Tim pointed to his home page for source code) to get access to XML from within a method. Tim explained that [WebMethod] attribute really does three very different things:

- dispatches SOAP request to the right method

- generates WSDL (using reflection)

- Marshal parameters

There?s a way to turn off WSDL generation. This is handy if you?re doing web services the ?right way? (WSDL first). There?s a way to turn off parameter marshalling as well. I think this may be handy if you?re dealing with some odd XML coming your way or need to optimize the way your handling XML. Once you turn off these two [WebMethods] features, you can still them to dispatch messages to methods.

It is nice to know you can access raw XML in .NET WS, but I?m not convinced one would need to go down to this level without a really good reason.

WS DevCon: Keith Ballinger: WS-Security support in .NET

Keith?s talk made me understand why we can?t just use SSL to secure Web Services. In a nutshell, it comes down to two arguments:

- There may not be a direct connection between client and server. Web services requests can travel though intermediaries who have to understand the routing instructions, but cannot be trusted with the contents of the payload itself. Using SSL would obscure such routing information and make it impossible for the intermediary to process the message.

- WS protocols must remain independent from HTTP. SSL secures HTTP traffic, but won?t help with other protocols that may be used to deliver a SOAP message.

Keith?s talk was a good intro to WSDK, but didn?t remove the need to read the documentation. It was pretty clear current incarnation of WSDK services will require developers to do quite a bit of hand-coding. It should get better with future versions.

WS DevCon: Clemens Vasters: Web Service Extensions for ASP.NET

Clemens presented his way of using SOAP extensions to insert all kinds of useful things into ASP.NET processing pipeline. I was very impressed by the elegance of the design and the care Clemens took in terms of delivering the best user experience.

Here?s what I understood: What he come up with is essentially an attribute-based approach towards extending a WebMethod. Want to add authentication? Just add [KerberosAuthentication] to the WebMethod. Want logging? Add another attribute.

Once the attribute is set, Clemens? code does a lot of plumbing ? modified WSDL; update *.config files, inserts the extension handler before and after your WebMethod call.

But that?s not all! On the client side, once the WSDL is used to build a proxy, the thing actually reads the WSDL, finds the SOAP extensions; generates the code using System.CodeDom.

Using this design newtelligence have implemented the support for WS-Security. The difference with WSDK is striking ? Clemen?s way is much simpler to use. KeithBa should definitely take note and consider adopting this design for WSDK.

(A few days after the conference Clemens has announced a wizard that generates soap extensions. Nice!)

WS DevCon: Rich Salz: What WS Needs to know about PKI

Rich gave a broad overview of crypto, PKI and XML security standards. I?ve learned that there?s much more to XML security than WS-Security.

Talking about digital signatures, Rich mentioned canonicalization ? the process of getting the one canonical XML string from an infinite set of different XML documents have only in insignificant differences (white space, order of attributes, etc). Rich pointed out that canonical XML is not the same thing as PVIS (post validated infoset). It seemed odd at a time, but now it makes sense. XML Schema defines PVIS in terms of data structure that satisfies a particular Schema. Since XML Schema was designed to define data structures for any kind of medium (not just XML) it cannot possibly prescript things like the order of XML attributes or the amount of white space between angle brackets.

2002-10-16T18:06:00.000-07:00

WS DevCon: Yasser Shohoud: The Right Way to Build Web services

Yesser delivered a clear message, true for anyone building component-oriented software ? nail the interfaces first, then think about internal component design. In Web Services world this practice is also applicable. One should start with WSDL first. Then design service internals using classes and data structures. Point taken.

There?s little support in ?developer tools? for this kind of workflow, so Yesser have called upon the audience to write their congressman and demand better ?tools?. By ?tools? Yasser really meant VS.NET and seemed to brush off the reminders from the audience that there are in fact other developer tools that do support WSDL-first workflow. Being firmly in VS.NET camp, I nevertheless wasn?t impressed with the way Yasser made VS.NET the center of the world of dev tools.

WS DevCon: David Seidel & Mark Ericson: Web Services Diagnostics

David and Mark gave some good tips of how to ease debugging web services.

Some of the tips:

- Use HTTP headers to transport logging messages

- Always compare cashed WSDL with the original to make sure that the other end is still using the same definition

They also gave a demo of SOAPscope (looks like very interesting product). Contrasting with the previous speaker, they gave generous information about other diagnostic tools out there, including their direct competitors.

Some of the tools I?ve noted: proxyTrace, xmldiff

WS DevCon: Andres Aguiar : DeKlarit Demo

DeKlarit seems like a truly useful product, especially for people who are can start an application with a brand new database schema (as oppose to build an application on top of existing database). The design of the product and the way it integrates in VS.NET is very well done as well. Detailed product review by Chris is here.

2002-10-16T17:02:00.000-07:00

WS DevCon: Don Box: What?s in a Name: Types and semantics in a web service universe

Brian has a good account of this talk, so won?t repeat it.

I find it hard to crystallize key messages of this talk. Don seemed to spend most of his slot making an elaborated preamble and we out of time before delivering the key message.

I think he wanted to explain how to achieve compatibility of types with XML Schema and WSDL. XML Schema allows in fact substitutions of one type with another ?compatible? type using substitutionGroup attribute. How to use it with WSDL isn?t clear to me.

2002-10-16T16:32:00.000-07:00

WS DevCon: Panel

Here are some bits of the panel discussion I had time to jot down (lots of detail missing).

Q: Thoughts on denial of service attack against Web Services?

A: (Noan, Tim, Clemens): There?s no simple answer. SOAP makes such attacks easier to do and harder to protect against. To some extent, DOS is better solved at the network infrastructure level. When the flood of request hits your SOAP server, its probably too late.

Q: Why does Microsoft has two SOAP stacks?

A: (Don, carefully choosing words) Msft will make it really really simple to choose one over the other. [Read this any way you like?]

Q: Any performance metrics that compare Axis against .NET SOAP?

A: (Glen, Don) Axis is much faster than Apache SOAP. .NET Remoting is the fastest tech for distributed apps.

Q: How valuable it is to become Web Services expert?

A: (Don, others) In 5 years WS experts will be as valuable as TCP/IP expects today. Unless you work for msft .net group or other ?plumbing shop? there will not be much opportunity to makes loads of money.

Q: Why don?t we see lots of commercial public WS out there?

A: (Don, Sam, others) MSN will have more consumer oriented with future releases (ex: mappoint) Most of the WS implementation will be done in the context of enterprise in-house developments, which in fact constitutes the majority of current and future software projects.

2002-10-16T15:53:00.000-07:00

WS DevCon: Christopher Dix: .NET, XSLT and Web Services

In my view, Chris has presented a truly original way to use XSLT and to build Web Services.

The idea is simple. Fundamentally, a Web Service is nothing more than a processor that takes XML input and generates XML output. XSLT is a language to define transformations of one XML document in to another. If one thinks about WS and XSML in these terms, they are indeed pretty close. As Don put it ?SOAP is XSLT with a longer wire?.

So what Chris did was to create a Web Service almost entirely in XSLT. I say ?almost? because you still need to write basic plumbing (HTTP listener, trigger XSLT) in ?regular? code.

The question is: how is it useful?

After his talk I?ve bounced an application idea with Chris and others. I started thinking about how to expose inference engines and/or expert systems as Web Services. This problem is coming from the product I?m working on. Any application that is based on inference engine (such as CLIPS) is defined in terms of ?facts? and ?rules? about those facts. If one can find a mapping between CLIPS rules and XSLT templates; as well as mapping between CLIPS facts and SOAP messages ? then you might have a generic WS interface to any inference engine application. That would be pretty cool. Now, I don?t pretend to know how to build such mapping. But I promised Chris I?ll spend some time thinking about it :)

2002-10-16T15:25:00.000-07:00

WS DevCon: Scott Seely: Using inheritance with WS

Scott has shown a design pattern that allows people to use class inheritance with classes that implement web services. I did not grasp all the fine details of the code and unfortunately I can?t find the code examples on Scott?s site. But from what I remember, he?d end up with a base class that would dispatch calls to sub-classes based on the type information found in WSDL. For many sub-classes you would end up with unmanageable switch-case construct. Then again, maybe I just didn?t get it?

2002-10-16T14:57:00.000-07:00

WS DevCon: Noah Mendelsohn: XML Schema

Apart giving an overview of the finer points of XML Schema language Noan made a very interesting point about the spec:

XML Schema is not just for XML documents. It was designed to be a generic language to describe data structures. So far, the usual use-case is to use it to define the structure of XML document. Noan made it very clear that nothing in the spec restricts its usage to describe data schemas held in other mediums such as in-memory representation, relational database, etc.

After the talk I?ve asked Noah is he knew if XML Schema is already being used beyond XML. I was specifically interested to know if databases would support XML Schema as a way to define the database schemas. Turns out, major DB players (DB2, SQL Server, Oracle) have fact started to work on it.

Memorable quote: ?There is no such thing as simple feature?

2002-10-15T14:59:00.000-07:00

WS DevCon: Steve Lougharn: When web services go bad

Steve was fabulous! He and Clemens share my prize for most entertaining speakers!

In a well-prepared presentation Steve showed what it takes to roll out a large-scale commercial web service. As expected, in large-scale systems web services don?t really make life easier. This stuff is hard by definition.

Steve offered advice regarding keeping healthy development environment for such projects such as:

- No barriers between r&d and operations

- Deploy early and often

- Operations tasks should be described as use cases

- Track operations issues as defects

- Instrument the code

Not surprisingly, most of the points are similar for a ?conventional? large-scale software project.

2002-10-15T14:33:00.000-07:00

WS DevCon:Peter Drayton: Designing RESTful SOAP API
Peter gave a great intro to REST design pattern and the philosophy behind it. I?ve learned there?s lively debate going on between REST and SOAP ?camps?.

Rather than fueling the debate, Peter crystallized the benefits REST that can be directly used in SOAP world:
- Model your system as a set of resources. (This way you?ll be able to address them and work with them independently of the root SOAP interface)
- Assign logical URLs to resources. (Practically speaking, use URI strings as oppose to oblique Ids to refer to things)
- Define schemas for resource representations. (Now that a resource can be addressed via a URI, clients will also need access to WSDL)
- Enable discoverability of resources. (Allow traversal of SOAP ?content?)
- Provide appropriate resource manipulation operations. (Define a small set of operations clients can assume are supported)

After his talk, Peter, Tim and Sam talked about a need to get a WSDL from a SOAP URI. In order to enable discovery of resources, one would need to get a WSDL that corresponds to a URI referring to a SOAP resource. Everyone seemed to agree that this is one of the key missing links and will make "REST guys happy".

2002-10-15T14:01:00.000-07:00

WS DevCon: Glen Daniels: Apache Axis
Apart from good info on history of Apache project and the reasons behind open-source nature of Axis Glen gave an introduction to Axis architecture.

It sounded like Axis has a simple and elegant pipeline-processing type architecture where a SOAP message flows though a series of ?handlers?. Handlers can modify the message as it flows though and can also add or get name/value pairs (or properties) in order to communicate with other handlers. The whole thing is pretty extensible meaning that users can create their own handlers and plug them in various places in the pipeline.

Amongst open issues Glen talked about difficulty in describing a SOAP error (fault). So I?ve wondered if there?s a way to marshal exception object within SOAP?s fault code? The simple answer would be to XML-serialize exception object on the server, pack it in fault code and re-throw it on the client side. That led to a hallway conversation with Steve Loughran who kindly answered by essentially saying ?It?s more complicated than you think!?. Steve pointed out that such technique will not really interoperate because CLR function call does not prescript the kind of exception it could throw. So .NET web service can potentially throw any kind of exception, while client-side Java code must know in advance which exception can be thrown by a function (via ?throws? keyword). There seem to be nothing in WSDL to describe how the error would be reported. Looking forward to learning how to do this right.

2002-10-15T13:51:00.000-07:00

WS DevCon: Sam Ruby: ?Interop is All?
Chris and Tim could not have picked a more appropriate keynote ? the ideas from Sam?s talk resonated in most of the presentations that followed.

The key take-away: Do NOT use SOAP is you always control both ends of the wire in a distributed system. The ONLY reason you want to use SOAP is to assure interoperability with other SOAP end-points (web services and clients).

Sam points out that systems that use things like ADO over SOAP are not interoperable right now, thus should not be used.

I had a pleasure to discussed .NET Remoting with Ingo in this context. People who think about building distributed applications should really think if they really SOAP (and consider Remoting instead). And inversely, if you need interop ? don?t use Remoting?s SOAP channel. It really gives people false impression they are building Web Services while they are using slow protocol inside a system that isn?t built for interop.

2002-10-15T13:41:00.001-07:00

Back from WS DevCon.

This conference blew me away. It was at once exciting and humbling to spend two full days in the company of real experts. The energy in the room was truly amazing ? there was zero marketing bs, no Java-vs-CLR religion, the audience was engaged in every talk, complementing it with questions and comments.

I won?t try to summarize the talks here ? Brian Jepson posted near real-time account on his weblog. I?ll simply add my own impressions and observations.

The number one reason you?ve been hacking too many web services?

You no longer see the angle brackets, "just blond, brunette, redhead"

2002-10-15T13:41:00.000-07:00

Back from WS DevCon.

I won?t try to summarize the talks here ? Brian Jepson posted near real-time account on his weblog. I?ll simply add my own impressions and observations.

The number one reason you?ve been hacking too many web services?

You no longer see the angle brackets, "just blond, brunette, redhead"

2002-10-08T14:18:00.000-07:00

Next Stop, "Greenwich": Enterprise IM Takes Shape with Platform Roadmap [Microsoft PressPass]

Details on RTC platform from Marketing VP for Windows Server.

2002-10-08T11:52:00.000-07:00

I'm going to Web Services DevCon!

I've never been excited about the conference, the topic, the speakrs and of course, the attendees since SIGGRAPH '96!

I certanly hope to meet the people behind the weblogs I'm reading.

2002-10-07T07:03:00.000-07:00

Compuware has a .Net Profiler for free. It helps pinpointing performance bottlenecks in .NET applications. What have you to loose ? FMS Inc's new Total Net Analyzer is here and this is a terrific tool to complement code-review sessions with your peers. It's more thorough than the VS.Net compiler (both C# & VB) and costs ~$500.[SBC's Radio Weblog]

Another good addition to .NET development environment to keep it healthy. FMS' Total Net Analyzer along a free Assembly analyzer found on Gotdotnet should really be embedded into round-the-clock build scripts. I would advocate using such tools even before the code review. Code analysis reports should be as visible as compilation warnings, because that's precisely what they are.

2002-10-04T07:59:00.000-07:00

Peter Drayton lists an impressive number of talks by some of the greatest minds of computer industry. The archive contains hours of video-on-demand. You'll want to find time to check some of those talks!

Food for Thought. The recent pilot of MIT's OpenCourseWare reminded me of another brain-food goldmine that I ran across a while back: the Multi University/Research Laboratory Seminar Series. Organized by a combination of Microsoft Research, Carnegie-Mellon, MIT, Stanford, University of Washington & Xerox PARC, the topics & speakers are absolutely bluechip. [...] [Peter Drayton's Radio Weblog]

2002-10-03T07:08:00.000-07:00

Greg Reinbecker and Sam Gentile are Announcing Groove Experiments shared space and work.

What a great idea! As I said before, I believe that the Groove Platform holds immense promise, way beyond the Groove Workspace and tools. When fully exposed to developers, Groove Platform has the potential to dramatically increase the value of applications enabling collaboration fro within the existing applications. (I hope I get admitted by Sam and Greg)

2002-09-27T08:40:00.000-07:00

Mail-to-weblog seems to be the only real way to contribute to weblog from different machines. I'm puzzled why Dave and Co still have not implemented the obvious functionality of being able to access weblog editing UI directly from http://radio.weblogs.com ???

2002-09-26T06:36:00.000-07:00

Microsoft is preparing a version of MSN Messenger with new features that will be available exclusively to paying subscribers of the MSN 8 online service. Joe Wilcox reports on CNET.

One of the major MSN Messenger 5 features available exclusively to MSN 8 subscribers are enhanced parental controls, which can be applied to online service client and e-mail features.

MSN Messenger 5 [free XP-only version] offers a feature called "Browse the Web together" that was not activated in the beta. The client also offers access to the user's history of .Net Alerts, which serve up traffic, stock, auction and other tracking information in IM. Furthermore, the contact search appears in MSN Messenger 5 but not in Windows Messenger.

2002-09-24T07:15:00.000-07:00

Head of Microsoft Research Rick Rashid talks about ongoing MSR projects.

"Sideshow" is the internal name for a project in which the company has developed an application that displays a series of windows with useful information on a user's desktop. Using XML and Microsoft's .Net Web services technology, Sideshow can reach out to the Web, corporate servers, or the computer's hard drive and provide quick views of data relevant to the user.

Sideshow team has published the project paper last month that described "notification and awareness platform". It looks like intelligent dashboard that apparently is being regularly used internally at Microsoft by 7000 users. Integrated in Office, this kind of tool will represent dramatic evolution of personal dashboard. I predict Office people are or will be working on productizing this. Groove team should definetely take note.

2002-09-23T11:24:00.000-07:00

Cathleen Moore reports from IM Planet Conference. These seems to be a consensus amongst the players (Microsoft. Lotus, Groove) on the protocol.

In the end, rather than a feature/function bake-off among vendors, the real value differentiating corporate IM will be the ease with which enterprises can extend IM and presence awareness into applications and infrastructure, [Lotus'] Dies said.

2002-09-21T19:44:00.000-07:00

Wanted: software designers who don't know software.

Thoughts on how to build future software.

2002-09-19T07:53:00.000-07:00

Execution is king. After reading this interview with Larry Bossidy (ex Chairmain/CEO of Honeywell) I feel a bit uncomfortable spending more than a few minutes writing those strategy notes. His book may not be getting stellar reviews, but the interview is very refreshing.

2002-09-16T10:32:00.000-07:00

Steve Gillmor and Mark Jones are interviewing Charles Fitzgerald .NET Strategy guy.

"We're certainly interested in the next generation of tools for access and collaboration and giving people tools to actually do something with that information. And today's portal model certainly falls far short of where we need to be. ... There's a broader road map where I think over the long term you're going to see a hybrid model of [collaboration] services, and we're big, big fans of peer services obviously."

2002-09-16T08:47:00.000-07:00

Longhorn vs. Groove Platform, Next Office vs. Groove Desktop
The race is on.

Ever since I've learned about Groove I was amazed as to why such obviously useful technology has so little competition. I've explained it to myself as "Well, Ray and his team kick ass". Then, I've learned about that $51M deal and I though to myself - "Wow, Ray and his team really do kick ass! Not only they have great tech, but also Microsoft won't be killing them anytime soon. Perfect!"

Recent news about Windows group buying small P2P company didn't change my views of Groove's team of course, but comes as a sharp reminder of age-old truth about Microsoft - one should treat Microsoft's product groups as successful and highly competitive companies on their own. In order to succeed these "companies" sometimes step on each other's toes. They are opportunistic and may make business decisions that benefit their own group, but at the same time putting pressure on other groups. This rule is espetially true with cach-cows like Windows and Office.

What does this have to do with Groove? I'm guessing Office group (Jeff Raikes & Co.) was the primary proponent of Groove/Microsoft deal. They have good reasons to do so. As Office people were building their road-map, they desperately needed the technology that allowed cross-enterprise online/offline collaboration technology. I suspect they have talked to Windows group and inquired how Windows would address this requirement. At that time Windows people didn't have a good answer. By making the decision to invest in Groove, Office group got their roadmap in good shape. But it also have indirectly put the pressure on Windows group to come out with their answer as to how Windows itself would support the collaboration scenarios required by Office.

The acquisition of XDegrees by Windows' storage group illustrates that Windows group is now serious about providing cross-enterprise file sharing abilities. One can expect Longhorn to provide a lot of Groove-like technology. (One can also expect that Windows implementation will be built on GXA from the ground-up.)

What this all means for Groove? It will soon have competition - Windows. That's... serious competition. (Strictly speaking, Groove's basic productivity tools like calendar, project management kind-of competes with Office, but Groove's integration with Office will likely render those tools irrelevant for Office users). The good news is Longhorn is far away, so Groove still has time to mature its platform and get good client base.

The other good news is, Windows isn't after everything that Groove stands for. Groove has three crown jewels in its crown: Moving data cross-enterprise. Working online/offline transparently. Lastly, enabling collaboration features deep inside desktop applications. Windows Inc. wants the first two. The last one is still undisputed and has to be protected at all costs. So in order to sustain the business over the long term, Groove has to excel in enabling collaboration of desktop applications well beyond file sharing, well beyound Groove Desktop. This is Groove's chance to keep being one step ahead of its competitors.

2002-09-13T06:07:00.000-07:00

Following-up yesturday's post :- As CNet's Stephanie Olsen reports, big banks begin to pressure IM companies to interoperate. The fact that customers take matters into their own hands shows how software companies have really screwed up with IM. It seems that this time around software companies have a real incentive - paying customers wanting the solution. The common standard will be good news for Groove.

The question is, how much AOL, Microsoft and Yahoo will be charging to let others IM with people on their networks.

2002-09-12T14:31:00.000-07:00

Weblogs gave you the power of publishing.

Now Jay Link gives you O'Reily cover to go with it!

Make your own!

Thanks go to vowe.net

2002-09-12T13:53:00.000-07:00

Lee Finck is ready to give up on IM. He has all the right to. To me and to many other people passionate about software, it is painful to see the lack of technological progress in IM space. It is painful to see so many souls locked up in Socialist Republic of AOL behind the Iron Curtain of proprietary protocols.

That said, I don't believe that the absence of open protocol is the primary reason why there's no IM-based killer app yet, apart plain-text IM itself. Declaring that IM is in dead-end because of absence of interoperability is like saying that software is in dead-end because we have several OS on the market!

First off, the protocol issue is not unsurmountable. Look no further than Trillian for proof. If Trillian managed to unify access to all popular IM platforms, why others can't? Secondly, if someone has a great idea on how to use IM to solve a real problem, why wait for everyone to be on the same network? Why not target one single IM platform? MSN Messenger is open enough. Yes, IM interoperability will help successful application more successful. But it will not turn mediocre software into a overnight hit. Groove is of course a good example of a successful application that will greatly benefit from unified IM. But inversely, if Groove fails, no one can blame the falure on the fact that IM world isn't unified.

It is not how AOL doesn't give a damn I'm frustrated about. It is near complete absence of commercial applications that innovate around IM to solve real problems that I'm puzzled about. Maybe I'm just too impatient...

2002-09-11T08:34:00.000-07:00

Another addition to .NET developer's toolbox: Peter Drayton has a very useful little utility that allows provides automatic call tracing. Extremely easy to use and very practical for debussing those triky remoting or multithreaded pieces.

The TraceHook.NET service is a context attribute that provides automatic call tracing on attributed classes. It traces instance method calls & field/property accesses to the debug output, allowing one to monitor an application as it runs. The trace includes type names, method names, parameter names & values (both [in] & [out]), as well as any return codes or thrown exceptions. Full, commented source is included, thus it should also serve as an interesting demonstration of the use of context attributes & interception in the .NET platform.

The source code was based on beta release of .NET, so I've sent Peter a few code changes to make the tool work with the latest framework.

2002-09-06T10:53:00.001-07:00

This will be my next camera. That's when i'll quit developing film and focus on developing software!

2002-09-06T10:53:00.000-07:00

Blog bombing

In a bizarre surreal bow to the power of perception on the web, what you say about a page becomes just as important as the actual content of the page. [Uber]

Uber have introduced the term "google bombing". What i'm talking about here is really is a particular instance of unintentional google bombing. Any informaiton in public domain is a double-edge sword.

[TODO - to complete]

2002-09-06T06:57:00.000-07:00

Megnut's mom, who once was guest host on Meg's blog, has invented a new idea -- googlecooking. Meg says "My mother types whatever ingredients she has on hand into Google and then picks the most appealing recipe returned in the results." Smart! [Scripting News]

What a wonderful idea! So to define googlecooking in general terms, it is the process of searching the algorithms/resources needed to achieve all possible results using a set of ingridients and pre-conditions.

This might be far fetched, but I wonder how can this idea be applicable for things like supply chain management. Might be an opportunity for google-based applications for various verticals.

2002-09-05T08:04:00.000-07:00

Privacy vs. weblogs.

As people are creating webs of links between weblogs and establish common-interest communities, what are the consequences for privacy?

Preserving privacy means preserving the ability to own and control all of the personal information. It also means the ability to clearly understand what kind of personal information is being produced. From the first look at the issue, it would appear that we disclose exactly what we decide to post to weblog and not more. But as people begin to be part of weblog communities, another kind of personal information is being released. It becomes possible to determine the circle of friends for instance. It becomes possible to track patters behind individual's interests over time; to determine the level of interest of others to a given individual; to tell who's reading.

There's key difference between weblogs and newsgroups where most of the above bits of personal information was already in the public domain. With weblogs you can reliably answer the question Show me your friends and I'll tell you who you are. Also, with weblogs it is much easier to build analysis tools.

Let's imagine for a second that some guy Joe is an recovering alcoholic. Joe is not talking about this particular problem of his on the weblog. He want to keep this fact private. Now, Joe has friends whom he met in rehabilitation clinic. They choose to disclose the fact that they are recovering alcoholics. Once it a while Joe's friends link to Joe's weblog. Now imagine a tool that tracks links between weblogs over time and achieves statistically representative data sample. When it becomes clear that 10 of his friends have problems with alcohol, will Joe be able to keep his secret for long??? What if Joe decides to run for president?

I'm not trying to be comprehensive here and I'm sure anyone can cook-up another technology-assisted privacy invasive scheme. My point is, many people are not realizing what kind of personal information is really being released as we continue to build these webs of weblogs. Maybe its too early to ring a bell, but weblog tool vendors should definitely think about this (or talk to experts). It is very important to realize they are developing tools that create wealth of personal information and it should be their responsibility to allow users to own and control all of the personal information.

2002-09-04T12:34:00.000-07:00

Thanks to Jeroen Bekkers for encouraging me to switch to radio. The tool is very flexible. Switching could be made simpler though - couple of hours if you don't find the homepage template you like. I'll keep blogger-radio bridge running for a while.

My first feature requests:

Allow automatic import of the content from a different blog and post it with the right dates (in the past).
Pure Windows client for posting that would use MS Word for text entry (a-la Outlook)

2002-09-04T08:07:00.000-07:00

Just about every blog I’ve checked this morning had a link to Ozzie’s reply to Joel’s article on Platforms. The fact that a lot of people expected a public reply from Groove/Ozzie tells a lot about how blogs have changed high-tech PR game. Can’t wait for other software executives to catch up!

In my opinion Ray Ozzie did a wonderful job explaining Groove’s strategy. The bit I especially like is the appearance of previously unpublished info on Groove OEM Kit. The existence of such kit confirms that Groove does recognizes the demand for Groove runtime and is working towards making it generally available. This is good news.

But notice have Ozzie skillfully blends two things together. He calls Groove Workspace an application, but it is a platform too. Yet, there’s “Groove Platform OEM Edition” as well, but its not yet generally available.

Let’s see if I can get it straight. First of all, there’s Groove Platform. 3rd parties can develop applications on it using OEM Kit. Then, there’s Groove Workspace. It is an application that runs on top of Groove Platform, but also allows 3rd parties to extend it by writing Groove Workspace Tools.

The best analogy I can find is Windows and Office. Windows is a platform. Office is a Windows application, but you can also extend it by writing Office plug-ins and marcos.

Ok, this sounds very very reasonable and healthy. So what’s wrong with Groove’s picture? The priority. Clearly, Groove has made a conscious decision to invest much more in support for developers of Groove Workspace Tools while providing a bare minimum support for Groove Platform developers. Imagine if 95% of MSDN was allocated for information on how to develop Office plug-ins and 5% was given to help out Windows application developers. And you had to "negotiate" with Microsoft to get that 5%!

From the architectural perspective, there’s a risk that some of the critical functionality would creep into Groove Workspace from Groove Platform. Don’t get me wrong, I’m convinced that Groove’s architects are very smart and experienced people, but its really tough to build a nice generic platform when you have one single application that tests it.

I think Groove has made a strategic mistake, but its not too late to correct it. Groove needs to increase the priority of getting Groove OEM Kit into general availability. They also need to start encouraging people to develop on Groove Platform just like they encourage to develop on top of Groove Workspace. So far Groove have done a really good job listening to developers, so if enough people complain, I believe they will find the way to solve the problem.

2002-09-03T06:06:00.000-07:00

It's about time Groove has got some competition!

Matthew French reports on CADwire.net on Availl

From the first glance, it looks like Availl has solved pretty much the same problems that Groove runtime solves (security, firewall traversal, online/offline handling). In contrast with Groove Availl seems to be only focusing on selling the 'briefcase' solution - file sharing and synchronization. Compared to Groove, Availl's file sharing product looks extremely simple to setup and way more transparent - no shared spaces, no UI.

2002-09-01T14:39:00.000-07:00

Joel Spolsky has some harsh criticism for Groove strategy. While I disagree with most points expressed in the article, I think the core message is consistent with the one expressed by me here and Philip King on Groove developer's forums. If this sounds odd, read on.

When talking about multi-layer product such as Groove, one should define the terms to avoid confusion. So first, a quick terminology intro. "Groove Runtime" defines a UI-less piece of code that offers such services like secure communication, firewall traversal, persistence model that allows efficient broadcast of data set changes, handling of online/offline contexts, etc. "Groove Transceiver" is an application that uses Groove Runtime to offer the UI for such features as shared space management, user management, instant messaging etc. "Groove Tools" are hosted inside the Transceiver and allow the user to perform a specific function inside Groove Transceiver.

Joel claims that Groove Inc. is making it hard to build applications (Tools) on Groove. I think this statement simply cannot be further from the truth. Clearly, Joel have never tried to create a Groove Tool with VS.NET Toolkit. For a young startup it is truly impressive to see how much effort was put to simply the creation of Groove tools.

Joel also claims that Groove Inc. doesn't know that Groove is a platform. I wonder why is there a developer's zone on Groove's site? Again, this argument doesn't stand basic analysis.

So do I think Joel's article is wrong? No, it simply fails to capture the real issue with Groove - there are two different views of "Groove Platform". One is being offered by Groove Inc. and another is wanted by ISVs.

Groove Inc. defines the platform as Runtime plus Transceiver. Want to build on the Groove Platform - write a Groove Tool. Most ISVs define the platform as Runtime. Period. Most of the people thinking of integrating Groove's collaboration capabilities only want the Runtime. This is why Joel's analysis of cost/benefit of integrating Groove with CityDesk is right on the money - it illustrates the Groove/ISV gap.

In Groove's defense, they do claim that the Runtime can in fact be used by ISVs in their applications (whoever there seem to be technical issues with this model in the current release). But still, the primary positioning of the platform is Runtime+Tranciever.

So, its not that Groove thinks they don't have a platform (or they're clueless) - its just that their priorities are out of sync with those of their partners. This strategy flaw can be corrected. And if enough people complain, I believe it will.

2002-08-30T07:29:00.000-07:00

Having a discussion with Mark Smith from Virtual Methods started by comments about what kind of groove tools will succeed.

Mark: Obviously I 100% disagree with your comments about whether Groove Tool Vendors will succeed. The major problems I've got (apart from price/quality issues) with our larger competitors is that they treat software development as almost a solo activity, carried out in an environment tighlty bound to a developer IDE. Your link to the Infoworld Booch interview shows that other firms are pushing in the Analyst<->Designer<->Developer direction, as borne out by their most recent product releases. My view is that the other direction-Customer<->Analyst<->Project Manager -is being neglected. I also know from direct personal experience that a Groove space provides a better way of specifying and designing software than the current hotch-potch of email, Word documents, scribbled notes, meeting minutes and so forth. Would you agree that the integration of requirements, design and project management can only make softeare projects less painful?

It seems we’re talking about two different issues – first being whether Groove tool vendors will succeed and the second being the vision and the target market of UMLTool. Obviously, its not my place to critique UMLTool and Mark's vision behind it. I sincerely wish Virtual Methods luck making it a hit! I’ve only used UMLTool as an example of a new application that knows collaboration in an established market of feature-rich tools that don’t know collaboration. Fair?
I think we’re in agreement that current software development tools simply lack collaboration feature set. Yes, software development is highly collaborative process that involves many different types of stakeholders as Mark points out. I would pay to have the better collaboration tools for software developers. But I would only pay if those tools don’t take away my other favorite features. In my work I use UML using Visio. I’ve selected this tool because it is feature rich, integrates with office, stable, gets the job done, blah blah blah. But this tool doesn’t know collaboration. So naturally, I turn to UMLTool. Will I end of switching from Visio to UMLTool? I will, only if UMLTool doesn’t take away all of my other favorite features present in Visio.
I believe this very logic will be used by majority of buyers of Groove Tools. If users find groove tools sub-standard in areas other than collaboration, I believe they will opt to use their current tools and simply end up using Groove for file sharing. Same logic will apply to Groove-created tools like calendar. What would Outlook user prefer while arranging a meeting with a partner: seeing their shared calendar directly in Outlook or switching to different UI of Groove, going in the shared calendar and trying to figure out the conflicts with his/hers personal Outlook calendar?
Would you agree that sub-standard tools that know collaboration are less valuable to the user than full-featured tools with added collaboration features?

2002-08-27T14:21:00.000-07:00

While looking for info in groove integration exampes I stumbled on this post:

Date Posted: May/10/2002 11:02 AM

Posted By: philking (Member)

Moderator,

Also checked the controller and noted that is shutdown fine there. The sample controller code is too simplistic to use as any sort of meaningful test. Note that opening tools as per the sample doesn't allow the tools to work at all!

If GN is serious about selling Groove for controller use then that comittment should be evident through the samples and adequate beta testing in the contoller environment. This would also be an excellent test for the API classes which like any good class structure need to work independently of any UI or Tranceiver specifics.

Controller apps could end up representing a huge slice of the total user base of Groove technology. Currently the transceiver and standard UI components are far too limiting for powerful app developent. This shows through all tools currently available and the time, effort and cost to produce such mediocre functionality. The end result is that users come away less than impressed with Groove. Indeed many are reluctant to use it because the interface and functionality is far inferior to which users are accustomed in most windows apps these days. To understand why, one can simply look at the ListView component. It's functionality is so crude it wouldn't register on a feature comparison chart up against mature ActiveX grid controls on the market today. Sure we can sort of wrap ActiveX in tools but that does NOT deliver the performance or functionality of a component designed to be hosted in an ActiveX container totally integrated into COM.

Personally I believe developing Groove apps in a mature environement supporting native ActiveX containment is the way to go. Rich and diverse functionality can be delivered at RAD speeds providing the API doesn't choke developers efforts along the way.

GN marketers need to remember that no matter what hosts Groove technology, the Tranceiver or a Controller, a Groove license sale is a SALE! Controllers need more attention. Please bring this crucial issue up with GN management and help Groove proliferate to every desktop.

Still with grand hopes for Groove ...
Phil

05/10/2002 11:18:34

This post resonates with my concerns about groove's busines model. As much as I'm excited about groove's technology, collaboration should remain a feature of a desktop application. Collaboration framework/platrofm like groove must avoid making ISVs re-write their existing applications just to enable collaborative workflows. Even less groove should try to re-write those apps themselves!

I'm sure we all want to see superb collaboration functionality in something like Adobe Photoshop to take one example. Does groove seriously beleive that adobe will re-write their system to be hosted inside groove's UI?

What's more worrisome is the fact that some of groove partners seem to buy this strategy. They seem to beleive that groove is their golden opportunity to leapfrog established competition.
Let me pick on Virtual Methods UML Tool. Forget collaboration for a second. How does the UML modeling funtionality of this editor compares with competition (Visio, Rational Rose)? I think UMLTool is doomed because of very strong compatition it faces.

The bottom line: Fantastic technology like groove will not cause people to abandon their favorite tools. Groove should pay more attention to integration into existing applications.

2002-08-26T13:58:00.000-07:00

Just read Understanding GXA by Don Box. Very nice brief (8 pages) overview of a gigantic undertaking of building infrastructure protocols for web services. For people interested in WS protocol, i definetely recomment reading this article before diving in detail of any specific any specific protocol.

2002-08-26T08:00:00.000-07:00

"Spend A Day with .NET" Coding Contest
What a wonderful idea by Chris Sells. So I still have 3 days to decide if I want to participate!

2002-08-26T07:43:00.000-07:00

Paul Krill announces that Microsoft readies specifications compliance kit for Web services. It seems that with those tools in place the stage is set to build entirely standards-based groove clone. That would be a secure (WS-Security) data sharing (WS-Attachments) that can work though firewalls (WS-Routing). Of course this is highly speculative. Need to learn more about this...

2002-08-26T07:32:00.000-07:00

Brian Fonseca writes about application-level firewalls.
My prediction: there's absolutely no choice for firewall vendors to go in this direction. Web Services would represent a major target for attack. The richer services become, the more attack-prone they will be. The pain will only start with 'corporate' web services, but sprread on personal workstations as people employ more and more P2P tools.

2002-08-22T14:15:00.000-07:00

High-Tech Ventures by Gordon Bell

After 11 years this book remains a great resource for everyone who evaluates a starup's health or is thinking about starting a business. The book contains a very comprehensive set of really good questions that evaluate the business in every dimention critical to its success. The success is assured if you have the right answers to all of the test questions presented in the book. Great resource for high-tech enterpreneurs, VCs, hires.

Word of caution though. This book isn't really about how to build a successfull business. It's more about how to tell if a given business (or a business decision) is to be successful. Most of the case studies in the book are really annotated examples of falures, as oppose to examples to follow

Highly recommended.

2002-08-22T08:53:00.000-07:00

Assuming that Rational is hard at work on the stuff Grady Booch sees in the future stay tuned for some great developer tools! Just what i think every developer really needs.
"The idea of bringing the UML as a higher-level language that transcends most of our textual languages and enables us to do both very good code generation as well as reverse engineering and perhaps even [brings us] to the point of direct executability in some UML models." [Booch]

2002-08-21T13:08:00.000-07:00

Tom Sullivan writes about UML 2.0 spec.
Looks like they are adding more riguor and fine details to the already-too-detailed language. In years of informally using UML I've never seen people actually being successfull of using the language as it was originally intended - formal definition of a software system. Apart informal brainstorming, another great use of UML is communicating design patterns.
Where I've seen UML fall flat is when people attempt to capture an entire architecture in a static UML model. Such model will inevitabely go obsolete and out of data with the code.
IMHO what developers miss is some real practical tools that use UML beyound its informal whiteboard usage. It would be great to have a lightweight way to link the code and the UML model. Once the two are linked, ether the code or the model can be edited.

2002-08-21T11:14:00.000-07:00

Jon Udell's blog

2002-08-21T11:00:00.000-07:00

Jon Udell writes about RSS aggregators. If the phenomenon of blogging continues, soon many people will come to realize that they can use some automation of their daily routine of reading other people's blogs. Also, more and more people will "get" the idea of syndicating news posts.
We're clearly at the very beginning of this development. The tools like Aggie are pretty primitive, so there's lots of opportunities outthere. Email, IM, blog writing, news reading will come together in a few years in a very powerful way and the discuvery of RSS is a good step in that direction.

2002-08-14T11:57:00.000-07:00

Today's essay by Ray Ozzie has led me to his personal blog.
In an excellent note called Architecture Matters: The Rebirth of Public Discussion he argues that blog design pattern naturally solves the signal-to-noise problem of other means of online communication such as newsgroups and other similar services. This note has led me to thinking about "blog design pattern":
While I'm facinated by the phenomenon of weblogs I currently see a few little bit of an issue with blog design. In order to conduct an effective and fruiful dialog online, all parties have to be notified when a reply to the original post has become available. That's the basis of opration of mailing list, newsgroups and other 'conventional' methods of online communication. When a conversation is spread across many different blogs, it is difficult to see how a constructive conversation can be conducted when an author isn't notified that someone is replying to his or her post. It would not terribly difficult to build tools to solve this issue, but I'm not aware of any existing today.
Secondly, vast majority of blogs are personal. And a persone usually has to deside which ones to follow. This consequence of blog design pattern can easily lead to ideas being burried within a closed circle of friends.