Hi All,

I'm afraid I'm going to have to postpone the Architecture chat until next week, due to other commitments on Thursday.

So it's now rescheduled for next Thursday - 11^th September - if anyone has any topics, or saw anything cool at code camp or teched this week then leave a comment on this post and I'll add to the list of topics for next week.

It's been a back and forth discussion on the OAuth list for a while now - and with some people (zealots? ;o) having ideas they weren't even willing to contribute to the OAuth group until the IPR was sorted I'm glad to finally see it's been completed, this should hopefully help to improve the longevity and adoption of the standard by some who have been fence sitting.

So what's happened?

Well all all parties involved in building the original spec have signed an agreement of non-assertion, so now OAuth can be safely implemented anywhere without concern about lawsuits related to the IP in the spec.

OAuth is a pretty elementary standard in it's version 1 state - so in some ways it was
inevitable that this would happen (or at least I thought so) - there wasn't much to gain by any of the contributors blocking the progress of it becoming an open standard - but it's involved a lot of work to get it there by all accounts, so full credit goes to all
involved!

For a detailed writeup check out the post from Eran Hammer-Lahav or the post on Read/write web.

Conspicuous by it's absence is Microsoft, but for no other reason than they did not contribute to the OAuth standard - and so didn't have to sign - but of course LiveId does tackle delegated authentication - so in some ways they have a competing platform for handling delegation, presumably because OAuth doesn't provide a rich enough set of features at this stage to handle some of the more complex scenarios around scalability, signing message bodies etc. - though I'm just hazarding a guess, LiveId was presented
at the OAuth summit earlier this year.

Only just catching up on my blog reading from the last week or so this evening - so only just came across this, but  its pretty cool! - it's like deep zoom for power point - the
project is  pptPlex from the OfficeLabs team.

I wonder if anyone at teched / codecamp in New Zealand this year will be using it... I normally do a mind map when I'm structuring a presentation, proposal for a client/whatever - be great to actually present the mind map itself then zoom in to show the details...mmm

Hi all, there's an Architecture chat this Thursday - 21st August @ 11:30am onwards.

Some things that I've caught my eye since last time:

• .Net 3.5 SP1 and VS2008 RTM are here
• VS2K8SP1 and SQL2K8 installs are a big mixed bag, some people no problem, other tearing hair out!
• And so are the war stories such as regressions and critical changes - doh!
• Visibility and trouble with non-developers being unable to quantify the quality of developer outputs - sparked off by this post by Casey.
• Resharper 4.01 RC1 is out.
• ASP.NET MVC now and in the future (i.e. where it's going, and perhaps a discussion on future of monorail vs. MVC and Monorail 2??).
• Security practices on MVC, WCF etc.
• Redmine - an interesting (and nicer) alternative to Trac - wonder if there's an easy transition path for existing Trac sites?

Edit: updated this list with some suggestions from others.

If anyone has any topic suggestions - just make a post on this entry, or send me an email / IM message.

Details of previous posts and directions etc. can be found here on the wiki.

Here's the write-up for Architecture Chat #32 (from Thursday 7^th August 2008) ... some of the things we discussed:

• Jabl / Jass  (Javascript abstraction language) - we discussed this approach of writing languages that transform into another language - the tradeoffs of debug experience vs. productivity etc.
• Debug experiences in DSL's and in particular the lack of support in these language rewriting/transformation projects.
• The magic bullet language i.e. two-way debugging experience, language independent refactoring support, modular/pluggable language support (sorta like Boo).
• Code camp at the end of this month.
• Self explaining code / blaming code (i.e. code that can explain the decisions it makes in english, or at least identify the blame (i.e. steps) that lead to the result it selected.  Though we didn't bring it up, I quite like the way Rhino Security does this.
• xUnit thoughts after a month of using it on a commercial project.
• Ice ZeroC - WCF alternative?
• Sql Server partitioned tables, and the simpler query plans / improvements in 2008.

See the wiki for write up's of previous chats, thanks to all those who attended.