October 2005 Archives

IP Babble Banner

| No TrackBacks

The new IP Babble banner is made up of the following:

It ought to be obvious but the cloud is the IP network and, of course, it is babbling.

In the background I've included a picture I took while in Tokyo recently. It was taken from the top of the Hotel New Otani. Actually the building in which IONA has its office is just right and below of center behind the "sizeof(" code. The orignial (though resized) picture is found here.

The code layered over the photo includes some Web Services Description Language (WSDL) taken, but modified, from a sample WSDL in the open source ESB Celtix. It also includes some IP code from the ip.c file of the open source Unix project OpenBSD.

Big thanks to Barry O'Mahony for all the help with the website and the new banner. Barry's own blog can be found here.

Security and SOA

| No TrackBacks

In an upcoming SOA Executive Forum, in New York, Jon Udell will be spending a day discussing various topics including security. And he's soliciting input.

Jon's talk will discuss several security topics including gateways and WS-* standards etc. I sent Jon the following suggestion:

Regarding the upcoming SOA Executive Forum. I see that you will be discussing security and SOA. I didn't see in the summary paragraphs the issue of cross platform security in an SOA.

Modern SOA implementations in large organizations are likely to include multiple platforms: mainframes, Java, CORBA, .NET, COM, MOM, EAI, etc. With even a small mix of the above technologies a SOA may have to facilitate the propagation of security contexts/credentials from one technology/platform to another. It would be useful if your talk discussed this topic, perhaps by use of a scenario where a .Net client signs on and needs to be authenticated and authorized right through a J2EE based servlet or bean and on to a backend mainframe based CICS transaction. (Transaction propagation is for another day).

An Enterprise Services Bus (ESB) (I haven't decided whether I like this term or not) is supposed to address this problem. Well at least an Extensible ESB, as we would claim at IONA1.


1For disclosure purposes please note I work at IONA Technologies which builds an extensible ESB product called Artix. The views expressed in this article and others on www.ipbabble.com are my own and do not necessarily represent the views of my employer.

UPDATE  10/09/08 - I no longer work for IONA. I now work for Red Hat

Google To Attack eBay

| No TrackBacks

Well it seems if the blogsphere is to be believed Google is about to launch it's attack on eBay.

What this article didn't speculate about is how this will also compete with blog hosting and other hosting services.

In a world where Google is buying up everything, taking over your desktop and partnering with the likes of Sun; where Apple has the most relevant hip consumer product on the market and the attention of the masses. Is Microsoft just fading into the sunset?

Apple's winner iPod is not only helping the top and bottom line but is also a serious marketing tool for it's other products. More and more people are looking at Apple for their desktops again.

And about the only thing that's hot on a Windows PC is Google. Google maps, Google News, Google just-about-everything, including your desktop.

Heck, even eBay is showing more signs of innovation with their purchase of Skype.

About the only relevant news from Redmond is that they are releasing an OS that few people really care about as far as I can tell. For one reason, because they've waited so long for long finger .. I mean Longhorn. For another they know deep in their hearts that it's really going to be nothing new - just a "new" and "improved" Windows. Perhaps I'm wrong. But is we've fixed the security bugs in our product but want you to buy it as a new product really something to get excited about? - Provided, of course, that they do fix the problems.

In the coming world of increased IP Babble and innovation I just don't see anything exciting about Microsoft. And what's really scary is that they don't seem to care or are at least not showing any signs that they do. It's like watching a hurricane slowly heading towards New Orleans.

Perhaps I'm out of touch with the Microsoft world of innovation. Please enlighten me.

4G to bring more babble

| No TrackBacks
Great Wired 4G Article in Wired last week describing some of the 4G initiatives coming.

The article discusses some of the challenges we face with 3G technology, some of the promises of 4G but also some of the 4G challenges. What's very interesting is the competing technologies. You'd think that we'd have learned the value of open architectures and technologies by now and we would expect some convergence in this area but it seems there will be lots of competing technologies based on different standards and proprietary technologies

At the same time you'd also expect that we'd have learned at this point that open standards are not always in the best interest of large platform players and that they want to try and lock people in. So we should expect the divergence in technologies.

So the consumer will end up with lots of choices but also with lots of headaches based on those choices. (Just look at the amount of hassle competing 3G technologies cause us).

And so we look forward to the 4G promise but are also weary of the all the competing technologies.

The next old thing - SOA

| No TrackBacks

This entry was inspired by the article The spiral staircase of SOA.

I find the latest SOA zeal both comforting and amusing. Several of us that were successful in building and deploying CORBA based distributed applications realized early on that using an OO methodology was generally doomed to failure. Many "failed" CORBA projects were because people tried to apply OOD to CORBA interfaces and the completely forgot about the fact that there was -or might be in the future - a WAN between their clients and servers or providers and consumers. Fine grained interfaces gave CORBA vendors a bad name and unfortunately there weren't enough "wizards" (smart developers) to go around. Service based interfaces forced designers to think more about essential contracts between their applications and those applications users/consumers.

I predict however that there will be similar failures in future SOA based projects - though perhaps less due to the emphasis on "service". The problem has not been the technology - for the last decade anyway (with CORBA and Java technologies like J2EE). The problem is that people continue to be poorly educated on best practices - on what a service is and why it's important to remember that though your application may "close" to it's consumer today it might not be tomorrow. Therefore consideration of a WAN is important. E.g. a good retail business understands that though their service might be exceptional they must also consider location. Location, location, location. Modern technologies do such a great job of location transparency that developers often program as if the service (in their view the "object") is right there in memory, close by.

That's why the word "Service" in SOA is so important and the word "Object" in CORBA was so dangerous. Hopefully people will take note.

Another thing to note based the article is the middleware fashion trend. I'll give you the short version - as a technology trend matures and gains wider usage, developers and architects usually encounter the complexities of enterprise features. As developers and architects struggle with these complexities someone else comes along with something new and innovative. This is not a bad thing. They also demonstrate how "easy" it is with a Hello World demo or something similar. Everyone applauds and jumps into the technology (for various reasons depending on who they are and their skill levels ;-). Then they hit the complexities of enterprise features, they get frustrated, and they start to look bad. And thank heavens for the new and improved technology that appears just when they need it (BTW have you seen its "Hello World" demo?).... and off the cycle goes again.

I understand that it is more complicated than that and that there have been great innovations. But there is merit in the argument that we are reinventing lots of stuff over and over again because we fail to master existing technology and improve it.

BTW thanks to Jon Udell for encouraging me to get this up on a blog. IP Babble was born.

Who is IPBabble

William Henry IPBabble is the personal blog of William Henry.

William has over 20 years experience in software development and distributed computing and holds a M.Sc from Dublin City University. He is currently working in the office of CTO at Red Hat, in the Emerging Technologies team. This weblog is not funded by Red Hat.

Posts are intended to express independent points of view, but understand that there is probably a bias based on the influence of working with standards based middleware for over a decade. (See disclaimer below)

January 2013

Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    

Disclaimer

The views expressed in this blog are solely the personal views of the author and DO NOT represent the views of his employer or any third party.

About this Archive

This page is an archive of entries from October 2005 listed from newest to oldest.

November 2005 is the next archive.

Find recent content on the main index or look in the archives to find all content.