[Klug-general] Talk at the next meet
macgyver at thedumbterminal.co.uk
Mon Nov 17 21:43:42 UTC 2008
On Saturday 15 Nov 2008, Karl Lattimer wrote:
> > > Either way, email sucks balls though... Some day soon we'll finally
> > > have the infrastructure in place to replace the email metaphor with
> > > something that actually makes sense... *sigh* to think xerox already
> > > solved this problem...
> > What is so broken with email? What would you replace it with?
> This is a pretty abstract problem. It's actually more about the content
> then the protocols.
> For instance, if someone sends me a meeting request via email, why do I
> choose to accept or reject it via my message viewer, why doesn't it
> appear in the context of my calendar?
> In the old days of offices (before computers) an inbox tray was
> something that items in the following contexts would be placed;
> * Things I've been asked to do
> * Things I'm doing
> * Things I'm about to do
> Now if we use this metaphor correctly in terms of email, we suddenly
> realise that email is absolute junk... All of it...
> So, think about how we organise mailing lists, building filters and
> folders to store all of those messages in relevant contexts, however its
> still not the correct way. Then take a look at how usenet/news groups
> work, Its kinda the same, but it works better in that context. Of
> course, news groups are categorised correctly anyway by their very
> Then of course, there's notifications, with things like facebook,
> notifications are becoming a large part of email traffic. However,
> notifications exist in a different context to letters, things I'm doing,
> things I'm about to do etc...
> So its pretty plain to see its all full of junk, and it needs some
> language processing, semantic linking and contextual separation at the
> natural points of separation rather than artificial ones. Moreover it's
> got to do all this without a user lifting a finger.
> Gmail/google calendar is pretty close to the eventual outcome, but it's
> still not there yet :/ And of course we want to leverage the power of
> "organise framework" to tie in snippets of data that arrive via email
> into the filesystem, linking together relevant data and automatically
> tracking the revisions of files that bounce about via email.
> There's a lot we can do to improve the situation, right now a huge dump
> of messages is not an effective way of organising things.
> There are also problems with point to point encryption too, rather than
> person to person... TLS for transmission should be standard by now but
> it's not universally supported. There are also a myriad of protocol
> inefficiencies such as imap having to download a whole message to
> forward it on, in this world of mobile devices forwarding on a 300Mb
> email via a phone is not plausible. pvanhoof is currently harping on
> about the protocol stuff, you never know IMAPv5 might end up being his
> Kent mailing list
> Kent at mailman.lug.org.uk
it took me a while to reply to this thread due to the content in the last few
messages (maybe a subject change would have been in order).
anyway this subject is interesting, do we all remember the promised winfs from
using sql server i think to store additional meta data with the files for
grouping and searching.
you could write something like this yourself with an open source database,
fuse and some bash scripts (sounds so simple).
i agree that email and other protocols has been used for more things that it
was designed for, like sending a digital photo to a friend i believe that
irc, msn are examples of better ways to do this as they are instant (FTP has
sort of died among non techies). another example would be web 2.0
technologies that bolt on to HTTP using various ways to overcome problems
such as statelessness, browser plugins such as flash used to generate fancy
webpages doing things that HTML can not.
We used to use email for tech support and customer service, but although this
was a simple tool, it soon became a nightmare due to archiving, spam and
searching, it ended up with people cc'ing others and duplicating
content. "ticket" tracking systems such as OTRS or RT are the way forward
Anything can be used as an hammer even a frozen banana.
Most users are not aware of the alternative ways of performing a task and dont
care as long as it works. i prefer open source as normally there are many
pieces of software that do the same task and i can just pick the one that
suits me. if you walk in to pc world then nearly all the computers will have
identical software, mind you these will all work in some way, just maybe not
the best way.
More information about the Kent