[Noisebridge-discuss] Virtualization resources

[Kragen Javier Sitaker]

On Tue, Jun 02, 2009 at 05:28:10PM -0700, Daniel C. Silverstein wrote:
> I certainly don't think you're insane, and, indeed I prefer free  
> software to proprietary software, but I don't think you're taking into  
> account the fact that time is not fungible.  Time spent setting up and  
> administering infrastructure is time not spent hacking on projects.

I think you mean "the fact that time is fungible"?  I think time is
partly fungible.  Time spent administering infrastructure requires a
different kind of thinking, possibly from different people, than any
particular project.

> I don't want to put words in Josh's mouth, but I suspect that he'd
> rather devote his time to hacking on muralizer than setting up and
> maintaining a git server so he uses a trusted third party such as
> github.

Right, and I understand and appreciate that. (And github is not
particularly "trusted" in the sense of "a security vulnerability"; git
is nice that way.)  It's just that I would hate for someone who wanted
to set up a locally-run calendar site to think that it was a waste of
time because we could just use Google Calendar.

> Again, if we can get the data out, which I'm pretty sure we can (vCal,  
> CalDAV anyone?), there's nothing shameful about this.  If you'd rather  
> see us running on a fully open stack, then by all means, make it  
> happen, but please don't knock folks who leverage open third party  
> applications to get things done.

If I had been trying to knock him, I wouldn't have started off by saying
what he did was "awesome".  Did that come across as sarcastic or
something?  Because it wasn't.

Josh writes:
> In this case, if we really cared about "owning our data," it's as
> simple as a cronjob that fetches the vcal file every day. ... And I
> don't have the problems of running my own backups or maintaining my
> own services, which full ownership would entail.

Yes, owning the data is easier to achieve than owning the service.  I
appreciate that you don't want to make the effort of owning the service;
I just wanted to articulate the point that owning the service is
valuable, even if you personally don't want to do it, because I read
your post as asserting that it was a waste of time unless it was better
than the Google version in every way.

> I'm glad people are gung-ho about Free everything, but, at the end of
> the day, I'd rather be creating cool things instead of maintaining our
> own copy of otherwise-commodity infrastructure.

I don't think they're mutually exclusive.  The fact that it's currently
easier to outsource your calendar to somebody else's server than to do
it yourself is kind of an accident.  It's an unfortunate and recent
accident.  One of the coolest things we could create would be a way for
it to be easier to create a distributed, decentralized calendar than a
centralized one and reverse that accident.

> On Tue, Jun 02, 2009 at 07:46:43PM -0400, Kragen Javier Sitaker wrote:
> > It's not *about* reducing overhead and managing risk and focusing on
> > core competencies and outsourcing.  Those concepts assume a kind of
> > fungibility of people's time that doesn't exist, a kind of central
> > management authority that doesn't exist, and a value system based on
> > profitability that isn't applicable.
> 
> Profitability?  Where is that coming from?  This is about maximizing
> awesome, which can be modelled like maximizing profit,

Overhead is an accounting concept, defined in terms of the relationship
between a cost and the firm's profits.  Outsourcing is a business term;
its purpose is to cut costs in order to improve profitability.  Focusing
on core competencies is something businesses do because it makes them
more efficient and therefore more profitable (except when the risks
strike).  That's where "profitability" comes from.

Efficiency is not the same thing as awesomeness.  

Modeling awesomeness as profit is badly misleading.

Kragen