good wrap-up from Anil Dash on "the new push"

'The controversy surrounding the Open Cloud Manifesto demonstrates the risk of trying to build interoperability behind closed doors and through exclusionary practices. Such environments are not conducive to building consensus, which is one of the key ingredients of successful standards.'

diffed. the explicit statement that there is no length limit to mail headers is potentially tricky

The newest rev to the Simple Mail Transfer Protocol (via fanf)

the newest rev to the Internet Message Format for email (via fanf)

'Based on case studies, this document identifies some of the factors influencing success and failure of protocol designs.' (via spicylinks)

wow, shit hits fan (via Aehso)

the OOXML fiasco. unsurprisingly, bureaucrats go to do whatever they wanted -- come on, what did people expect from the ISO?

interesting (via Michele)

a massive poster -- covers pretty much every technical aspect of the web, featuring the Towers of WS-Babel, Oxymoronic Intellectual Property, IP as nukes, the paved cowpaths of Microformats, the ruins of CORBA (represented as dolmens, Orbix fans) etc.

'SC34 suddenly has a lot of new ['participating member' countries] sending representatives, and I am interested to note the majority of their representatives are, as individuals, also Microsoft employees.' up to their old dirty tricks, then

'start by outlining the continuum of opinions, including the ridiculous or unthinkable. Then the narrower range of the reasonable; the Overton window. The job of the think tank is to move this, so that ideas that were once unthinkable become acceptable'

a concise, but complete, list of objections to Microsoft's proposed standardisation of the MS Word file format as an ISO standard. a strong case; tey even invent a new date representation, and treat 1900 as a leap year (MS vs. the Gregorian calendar!)

'Our justified love of open standards inadvertently promoted a straitjacketed approach to using XML. XML was felt to need a schema language.' ... 'The rise of agile programming techniques .. emphasizes the hazards of prematurely freezing a design.'

'Mailing list managers should be considered MUAs, i.e. that the result of list explosion should be considered a new message submission rather than anything like relaying or redirecting'. This is certainly the way DKIM and other standards are heading

an RFC for email autoresponders. If all the half-assed C/R mail-filter and "out-of-office" autoreply coders read this, my life would be so much easier

good article on the OASIS situation, explaining how "open standard" != "implementable by open source" necessarily, and discussing how to fix this