Re: IKVM.NET: Building a Java VM on the .NET Framework

On 2012-04-09 17:30, Jeroen Frijters wrote:
> If you're going to make the analogy, at least complete it. It's more like
> masturbating and then donating to the sperm bank, there at least a chance
> that something good might come from it (and you might not even know about
> it) :-)

I didn't want to complete the analogy, because I was thinking more along
the lines of my comment that I'd been working on projects mostly by
myself for 12 years. :P

> Yeah, that's the point where (as a techie) you discover that sales people do
> add value.

I grew up with a salesman (my Dad) so I've never really had any trouble
understanding that.

Although I'm having a real hard time resolving some of my feelings about
advertising (in general) and google (the company) and the fact that I've
signed up for an AdWords campaign.

On that. I rang the Google AdWords people this morning because they
offered me a "Free Consultation" as a part of my sign-up. Figured I'd
avail myself of some expert advice. What they told me was:

* I shouldn't have one campaign in 5 countries, I should have 5
campaigns in 1 country each.
* I should pay "Cost Per 1,000 Impressions" rather than "Cost Per Click".

I realised that by having 5 campaigns rather than 1 campaign I was
naturally inclined to spend more overall because you have a budget per
campaign.

And I realised that given that the click-through rate I was getting was
much lower than average (even outside one standard deviation) that by
paying for "Cost Per 1,000 Impressions" I was able to afford less
impressions overall.

So basically it was bullshit "up-sell" advice and didn't help me for
function in my interests at all. Pretty pissed off about that. And get
this... the other day when I set up my account, Google, one of the
biggest and allegedly most capable software companies in the world,
*lost my entire campaign* when their app crashed and after I'd credited
the account. It just "disappeared" when I logged back in after the
crash. I'd be upset, but it's actually a comfort to see how shit they
are. The bar isn't real high.

I realised while thinking about it for the first time this morning that
my preference would be to have really strong word-of-mouth brands, and
then I could tell people they save money by engaging me because I don't
charge them to advertise at them. :P

> At the same time, it's also easy to overestimate the value of
> your own solution. As an example, I plugged IKVM.Reflection a little bit in
> my talk and afterwards someone was interested in it and did some prototyping
> and discovered some missing functionality (I had not implemented case
> insensitive member lookup, because I though no sane person would use that).
> I'm now adding that functionality, but from that persons point of view
> they've taken a pretty significant risk (investing the time to write the
> prototype) with currently still unclear rewards. In my experience this
> happens a lot with software, so people are (rightfully) reticent to try
> something new.

Yeah. I'm one of those jaded cautious types now too. I have a software
aversion. :P

> A lot of NIH is actually rational behavior.

Well that's good to know, because forking other people's projects so I
can have control over what they do and how they operate seems to be a
theme of mine lately. I'm not sure if this has gone so far as to qualify
as a "problem" yet. I started my own club, wrote my own blowfish
encryption library, forked phpjs.org, started a new software company,
and on and on.

If you want something done properly you have to do it yourself. Right? :P