|
The fact nobody reinvented e-mail on top of a different set of standards is a good indication that the current stack is not as broken as some people would like to think. It reminds me of a previous life, at a previous time, when everybody reinvented the content management system based on slightly different approaches, because they thought existing CMSs were too big. It starts with serving pages, then you want multiple templates, then you want accessibility, then you want to authenticate with LDAP, then you want to use the LDAP groups as your groups, and then you need to tie groups to roles, because groups mean one thing to the org, and roles mean another just to the CMS, then you realize a relational database isn't that good a match, then you realize it would be convenient to attach some logic, and a document database is no longer a great idea, then you need multiple datastores for content, then you need new workflows on top, then... The devil is in the corner cases. They have been solved in different ways, by different parties, for decades, not always documented ("because it was just a simple thing, that once, and we fixed it"). Unless you've been there, you are doomed to make the same mistakes, and nobody was everywhere to avoid all mistakes. Why not dedicate an effort to migrate all the COBOL code currently running almost every banking transaction? That should be easy ;-) |
I've ... seen a lot of poorly-considered proposals to fix it.
The fact that email has one hell of a lot of lock-in does not mean that it's not-broken. Just that it happens to be in one hell of a local optimum from which reaching any preferable state has a lot of hill-climbing to do.
Much in the way that evolution is not teleological, and cannot identify some desired state it wishes to evolve toward, but must proceed, by stepwise incremental increases in fitness to another state, so it is with protocols and standards.
That's not to say that SMTP + fiddly bits doesn't solve some problems, occasionally elegantly. But the leap from there to "it's not as broken" is ... rather large.
I've read TFA and am going through comments to this thread. There are some ideas I like, some I question. I'd suggest that this proposal seems better though through than many others I've seen.