Simon Fell > Its just code
cool, Mark Baker started a blog, welcome Mark.
Wes is reading A Deepness in the Sky. A great read, I just finished reading A fire upon the deep, also excellent.
BlogToaster appears to have gone down well with the bloggers, its now upto number 6 on blogdex, and there's 147 people using the toaster as of now. As MSN has a limit of 150 for the buddy list, you'd better get in quick if you're not using it already!. I ought to be able to bring up another instance of toaster to get another 150 people going, I really need to get the code finished and packaged, so anyone can run there own BlogToaster hub.
Henrik Frystyk Nielsen announces a revised DIME specification, and a WS-Attachments specification [which replaces the SOAP over DIME specification], both are now sporting MSFT & IBM authors.
Clemens Vasters has released an experimental WS-Security implementation for ASP.NET
Thursday, June 27, 2002
Worth a look, the IPR statements for the XML Protocol Working Group [aka SOAP 1.2]
WSIL?. Timothy Appnel on WSIL. OK, so I must be missing something.... How's this different from the old DISCO again? Can anyone explain? Simon? [Commonality] The main difference is that the spec has both IBM & Microsoft's name on the cover.
WebServices.Org: "IBM donate key Web Services technologies to Apache" The Web Services Invocation Framework (WSIF) and WS-Inspection for Java Implementation (WSIL4J) are now Apache projects. [snellspace] Now all they need to do is fix the license on the dependencies, so its actually usable.
XMethods uses SonicMQ for asynch version of XSpace. Tony Hong from XMethods is here, showing a nifty evolution of XSpace. It's currently a service that implements a simple shared address book. The concept is that of a tuplespace updated and queried by SOAP messages. The new twist, slated to become available July 15, is the use of SonicMQ, a JMS (Java Messaging Service) provider, to add reliability, security, and pub-sub notification services to the message flow. ... [Jon's Radio]
Sunday, June 23, 2002
Moby Says Techie Fans = Fewer Sales [Slashdot: News for nerds, stuff that matters] Bollocks, if 18 was actually any better than previous offerings it might sell more. I was very disappointed with it [and the latest from DJ Shadow]. By far my favorite new music is Solid Steel from DJ Food & DK, absolutely awesome, hasn't been out of the walkman all week.
Saturday, June 22, 2002
Almost got day 10 done, but for some reason Mozilla and IE both leave a gap between the top and the start of the nav cells, It looks fine on Opera 6 though.
Microsoft takes hostages for next anti-Java legal round. New lawyers, please [The Register]
New Books. My copies of John Cough's "Compiling for the .NET Common Language Runtime" and Martin Fowler's "Refactoring" just arrived... Great! [Commonality] Good choice !, I've heard Peter say Cough's book is good, and Refactoring is a favorite of mine.
Wednesday, June 19, 2002
Day 8: Constructing meaningful page titles : Nothing too strenuous so far, I already had DOCTYPEs everywhere, and the blog already had the lang attribute, it took me all of 30 seconds to get the PocketSOAP site updated with a lang attribute. Just finished following Jake's instructions on how to add the date to the archive pages.
Editors' Newswire for 19 June, 2002. Newswire stories, including: Sun release Java Web Services Developer Pack 1.0. [xmlhack] Does this mean there'll be a flury of JAX-RPC / JAX-M toolkits to follow ?
Ben Lowery posts HttpCompressionModule, an IHttpModule implementation with source that does gzip/deflate HTTP compression. [Peter Drayton's Radio Weblog] Hmmm, alternatively you could just turn on the built in support for this.
Sun may get seat in Web services group. A proposal being considered by the Web Services Interoperability Organization could pave the way for Sun Microsystems to join on equal footing with rivals Microsoft and IBM. [CNET News.com]
Tuesday, June 18, 2002
Interop results udpated to include the WebMethods endpoint, and the group C Sun endpoint, and revised results for NuSOAP, Axis & Apache SOAP. Axis seems to be shaping up, just needs the mustUnderstand handling straightening out.
doc/literal/wrapped is being discussed over on SOAPBuilders. My biggest gripe is that in .NET wrapped mode is triggered by the putting the magic word "parameters" as the part name in the WSDL, so this is no different to there being a real indication that its RPC in the WSDL. This really should be a switch on the WSDL tools. To make things worse, it seems like other people are following Microsoft down the same rat hole.
Sunday, June 16, 2002
Now, I'm using Peter Drayton's RSS 0.91 XML schema which abides the Userland mandated five hundred character max. However, Radio itself does not seem to abide by this rule when emitting the description element. ..... [Drew's Blog] Radio generates RSS 0.92 not 0.91, here's the spec
Sam, its working now, right ?
Drew has some good suggestions for the BlogToaster, expect to seem them implemented soon.
I have a very beta version of blogToaster up and running, feel free to give it a go [you need MSN Messenger]. Add "firstname.lastname@example.org" as a new contact, and start a chat session, enter "add http://www.pocketsoap.com/weblog/" and hit enter, repeat with all the URL's of the weblogs you want to get notified about. enter "list" to see the list of URL's you've registered. When BlogToaster picks up a change from weblogs.com of a URL you've registered, it'll send you a message.
I think it's absolutely critical for someone to provide a news/mail bridge, even if it's a "for pay" service. I would pay a few dollars a month to have my mailing lists shunted through a news server, and empty out my inbox and its tremendous traffic. News is so much easier to weed through, and don't interrupt me during the day with the flapping bat wings. :) [The .NET Guy: Rants] I'm on an insane number of mailing lists at this point, but I run most of them through the mail 2 news gateway built into Mailtraq, and read them with Agent. It works much better than any other approach I've tried.
Tim Bray : PSVI considered harmful
Posted some updated interop results that include the results against the Sun endpoint, looks to be in pretty good shape.
Rich Salz email me to point out that WS-Security will let you do digital signatures, something you can't do with SSL. Thanks Rich
Tuesday, June 11, 2002
But part of the potential of WS-Security is to be able to have an authenticated (and possibly signed) transmission without having to encrypt the entire message. Important, I think, for both the point-to-point case and with intermediaries.[Greg Reinacker's Weblog] Certainly important for the intermediary case, not sure about the point-to-point case, the only thing it seems to buy you is a consistent approach with the approach required for intermediaries. Its going to cost you though at run time, I guess it'll be a while before a WS-Security implementation has been tuned to the level current SSL implementations have. SSL also has the advantage of being able to amortize the cost of establishing the session key over multiple message exchanges, via HTTP persistent connections.
I've updated the web services security story with some input from Justin Rudd. [Greg Reinacker's Weblog] Putting credentials in a SOAP header without an ecrypted channel is a waste of time, but if you have an encrypted channel, you might as well use the channel's authentication support. WS-Security only starts to make sense [much the same as SOAP] when you have intermediaries.
Sunday, June 09, 2002
Now, here is a concept for all you bloggers and surfers of the world: next weekend - turn of your PC. Leave it off for the whole weekend. Have a life. Visit your family. Get some exercise. Spend some time with human beings, face to face. Its a lot more interesting than staring at a screen all day. Trust me! [John Burkhardt] What, the whole weekend, all in one go, brrrrggghhhh, that's some cold turkey !
I've posted a response to Sam's "Is Com Interop Fundamentally Flawed?" article. Hopefully this provides a unique look at the problem that not many other developers have supplied. [Drew's Blog] Ahhh Java/COM those were the days :). Its a pitty that not everything from Java/COM made it to .NET, in particular in Java/COM java classes that were serializable automagically got a COM implementation of IPersistStream.
WSDL Wizard Beta 3 of the PocketSOAP WSDL Wizard 2.0 is now available for download.
ROTFL Forget about the SOAP serialization for a second, but this actually seems more like a CLR problem to me. The runtime allowed the value 5 to be stored in a fooEnum value type. Heck, I just changed the method to do this: return (fooEnum)5; and it worked too. Looks like the testing the integral value upon cast for validity within the Enum is not there. So in the end I suppose they could add extra checks at the serialization layer, but it looks like the CLR is at fault for even allowing this assignment to take place. Then again, enums are just an abstract concept intended for compile time type checking. So if you do naughty things like this, maybe the compiler ought to be warning you?
more on Enum
Forget about the SOAP serialization for a second, but this actually seems more like a CLR problem to me. The runtime allowed the value 5 to be stored in a fooEnum value type. Heck, I just changed the method to do this:
and it worked too. Looks like the testing the integral value upon cast for validity within the Enum is not there. So in the end I suppose they could add extra checks at the serialization layer, but it looks like the CLR is at fault for even allowing this assignment to take place. Then again, enums are just an abstract concept intended for compile time type checking. So if you do naughty things like this, maybe the compiler ought to be warning you?
Well the fact that the cast is required indicates that its wrong, but you want to do it anyway. Drew pointed out in a later email that the original case of fooEnum.red | fooEnum.green on an enum with the System.Flags attribute is more much subtle, and less likely to get caught. The compiler ought to stop you doing bitwise operators on enums without the Flags attribute.
Thanks to Fumiaki Yoshimatsu for emailing me to say that the System.Flags attribute should fix my issues with the enum below. I gave it a go, and it looks good, here’s the revised schema ASP.NET generates for fooEnum
<s:enumeration value="red" />
<s:enumeration value="blue" />
<s:enumeration value="green" />
Calling getSomeFoo, now results in a response of “red green” so far so good. If I send a SOAP request, then I can also send “red green” and have it mapped back to fooEnum.red | fooEnum.green, cool. I do appear to have spotted what looks like a bug, in that if you try sending “red green” with the HTTP GET binding, it chokes with an error.
Saturday, June 08, 2002
I've been looking into how to handle xsd:enumeration in PocketSOAP, I have something that works by mapping it to a VB enum type, and spitting out a custom serializer to get the correct string on the wire. I was looking into what ASP.NET does with enums and noticed that the bitfield stuff that .NET enums support can result in invalid messages being created by ASP.NET. I'm not entirely sure what it should do in this case, but generating a schema invalid message doesn't seem right to me, here's an example.
public enum fooEnum
red = 1,
blue = 2,
green = 4
public class foo
public fooEnum getSomeFoo()
return fooEnum.red | fooEnum.green ;
And here's the resulting schema fragment it generates
<:s:enumeration value="red" />
<:s:enumeration value="blue" />
<:s:enumeration value="green" />
This results in a response message containing the value 5, a result which isn't valid according to the schema generated for the WSDL. Perhaps it should generate an array of values ?
I just updated the main PocketSOAP RSS feed to RSS 1.0, let me know if you spot any problems.
Trying out the latest version of Aggie
There's a beta Win32 build of the HTTP/1.1 support for PocketSOAP available, if anyone wants to try it out.
If you're Interop'ing to Apache SOAP 2.3, you might want to be aware of this issue.
Sam points out that DIME vs SwA is about choice. Given that that the DIME stuff for .NET is a separate add-on, it should be possible to do the same with SwA, we don't need to wait for Keith to see the error of his ways :)
Web Services Interoperability panel at XML One/Web Service One. The XML Web Services One conference featured a panel of Web services experts from a variety of companies demonstrating interoperability in a basic scenario prepared for the conference. [xmlhack]
Thursday, June 06, 2002
Wasn't .NET suposed to stop this sort of thing ?
Looks like I ran into a strange edge case with the xmbus HTTP server.
W3C weighs in on Web services. Five months after reorganizing its Web services work, the influential standards body is releasing a trio of drafts related to the much-hyped trend. [CNET News.com] But will SOAP 1.2 ever make it to last call ?
Microsoft plans new Web services push. The software maker will announce plans Thursday for technology that will allow businesses to authenticate user identities between companies and applications using Web services standards. [CNET News.com]
Mark is upto all sorts of interesting things with referers and blogrolls
LOL, from the pocketSOAP.com refferer logs - http://referers/are/easy/to/fake/
Cool, I see Apache HTTPD on Windows can handle ISAPI Extensions
Can anyone clarify how HTTP/1.1 persistent connections and authentication interact ? I thought the authentication header would apply to that request only, but with IIS, it seems to authenticate the connection, subsequent requests to the same URI on the same connection don't need the authentication header.
There's now a FAQ for PocketSOAP
Security. The May 27, 2002 issue of eWeek contains four articles on web-services security. This one by Timothy Dyck illustrates the issues by getting into the differences between the competing standards, SAML and WS-Security. [Doug Kaye: Web Services Strategies]