I was recently talking to Lee from
Social Technologies gurus Headshift,
who made a compelling argument for the use of flexible, organic,
evolving taxonomies (folksonomies) in “the enterprise”
(big business), rather than the typical retrospective, pre-conceived
taxonomies that are prevalent in enterprise
document, customer and content management systems.
This got me thinking about other
aspects of the internet that are forking into two alternatives,
especially those where there is a traditional, or ‘assumed’ school of
thought and an interesting competing idea that doesn’t always suppose
the same restrictions.
Of course, things are rarely black and
white; most of these are not ‘either/or’ choices, and the internet
will continue to support these, and further choices. But ignoring
that, I’ve started to list some areas of internet technology where
it’s possible we could see a major technology/culture-shift towards
one of two opposite directions. I’d be interested to hear from
anyone who can comment on these, or can add further polarised topics
to discuss.
- Big apps vs Small apps
- This is another topic that Lee
raised. Whereas companies used to seek ‘all in one’ solutions
which could, for example, create all your data, manage it, keep it
under workflow and versioning, apply access control, publish it,
and give you statistics on its use, is there now an emerging trend
to use multiple pieces of simpler software that each delivers a
specific function? With more software publishing simple interfaces
that people can actually understand and use (e.g. REST over SOAP),
does this make the collection of smaller applications more
attractive? Or, alternatively, are companies fed-up with
supporting an increasing number of applications with differing
interfaces, processes and user accounts? Will there be a clear
winner, the niche providers or the big super-packages?
- This is another topic that Lee
- Local rich apps vs Online apps
- This debate is as old as the web,
but it’s no closer to being resolved. There has always been hype
around the central online hosting/provision of all your software
and services – to a point where you don’t need anything more
than a basic dumb terminal on your desktop, because it will only be
handling the interface. But common web browsers - after over 10
years of evolution - can still only render a relatively primitive
software interface, and bandwidth and latency have a significant
impact on many potential online software applications (e.g. for
music, video, or graphics). Will there be a shift back to
powerful, rich local applications?
- This debate is as old as the web,
- More control vs Less control (of personal information)
- This was just a quick thought I
had, and could be completely off-the-mark. There are many
occasions on which you need to enter your personal information into
a website (signing up for a forum, buying CDs or travel, social
networking services, and so on). Your personal information is
spread out all over the place, and there’s just no hope that you
can manage it all. It makes sense, to me, that this could move in
one of two directions. In the first, a centralised online service
provides a single, hosted location for your personal information
(possibly in FOAF format). It will probably manage more than just
this, but the important point is that this service can be used to
easily enter and synchronise your information with other
third-party websites. I see this as ‘less control’ –
certainly more efficient, but you will increasingly rely on this
(remote, not fully under your control) service to manage and
protect your personal information. An alternative option is to
retain the same ’single service’ idea that provides personal
information, but bring it back to your local computer. In other
words, have a P2P (or otherwise) local application that manages
your personal information. If anything or anyone wants to access
your information, it has to go through this (local) software, which
can be changed or deleted at any point.
- This was just a quick thought I
- Standardised Taxonomies vs Folksonomies
- Following on from earlier, are we
going to see the communal development of one or more ’standardised’
taxonomies? A personal hope of mine is (or, used to be…) that
Wikipedia will allow users to create and manage RDF-like
data/relations between entries; thereby making Wikipedia URLs/terms
an open, standardised (and inter-related) taxonomy of terms, usable
by anyone in their RDF (or other) metadata. RDF data in China
mentioning ‘Cats’ (the animal) could therefore be easily and
automatically mapped to RDF data in Mexico mentioning ‘Panthers’
(the animal). But could this ever work? Arguably, folksonomies
are currently so popular (and conversely, standardised global
taxonomies rare) because of the lack of (or necessity of) thinking
about the bigger picture, and fitting in with the work of others.
Folksonomies are built and evolved quickly because of this lack of
process and red-tape. Should we give up hope on a slow,
cumbersome, politically-heavy standardised taxonomy?
- Following on from earlier, are we
- Common Browser technologies vs Browser diversification
- The differences between web
browsers (platforms, products and versions) currently accounts for
34% of all web designers/developers nightmares (that’s just a rough
guess, but it can’t be far wrong). There is therefore an argument
that, like in the music and film industry, effort could be
concentrated and money saved if only a single browser/platform
existed. If Mr IE, Mr Moz, Mr Opera, Mr Safari and all the others
could get together around a table and square their differences,
developing for the web could be made so much more efficient. No
more CSS hacks! No more Javascript switches! Yes, you can rely on
SVG support and PNG transparency! But is that really likely?
Well, who knows – but it could also get worse. Like in the
early days of the browser wars, web browsers could increasingly add
new functionality and features that are only supported in a
specific browser, in an attempt to add value for users of that
browser. Through this, the browser could guarantee its success by
providing useful additions (which is why Netscape — who added
width and height attributes for images — is now so popular, and IE
– who added <marquee> — is nowhere to be seen).
- The differences between web
- Explicit Metadata vs Implicit Metadata
- We don’t really have much
metadata on the web at the moment. That’s not to say we don’t have
the data, we just don’t usually expose it on the internet. If the
Semantic Web is to happen, then we need for a new layer of
data/metadata to be published on the internet, underneath the
current articles, CDs and images that already exist. With this new
web of data at our disposal, a whole new world of opportunities
open up. But is this really likely? What if we find that the
trust issues (or quality issues) can’t be resolved? Maybe we don’t
need any explicit metadata at all – Google seems to cope with
measuring importance through implicit metadata (links between
pages), so could we not take this further? Can we further develop
the natural language processing algorithms and image processing
algorithms to deduce meaning and all kinds of other metadata from
existing content?
- We don’t really have much
- Death of Email vs Re-birth of Email
- I’ve been foolish at many points
in my life, but none more so than publishing my email address on
the internet since about 1992 (who was to know that Usenet postings
would still be available now?). This probably explains why I get
about 800 spam emails a day (thankfully, my spam filter cuts this
down to only a dozen or so). But I know of some people who have
given up on email altogether. Email is drowning in unsolicited
messages, it’s too easy to ignore, and did they really receive that
email? Has it been caught by their spam filter? So, could email
be on its way out? Or will it be given a new lease of life through
the injection of digital signatures, encryption, and new
technologies that let us ensure delivery, receipt, and limit
maliciousness?
- I’ve been foolish at many points
Do you see any of these choices being more likely that the other? Can you add any more discussion points?
