dawn wrote:
>I just heard a
>talk (at DEVCON) by a marketing person who had helped turn around a
>software company by having them remove the term "search" from any
>descriptions of their product. Everyone knows who owns the search
>market.
Hmmm, this reminds me of MySpace. It has basic features with a lousy
implementation, lots of bugs, and it's mis-used and abused by a large
segment of its audience who doesn't know or care much about software
quality. There's lots of room for improvement there, providing better
value without making it more complicated.
The traditional search engine is actually pretty poor at delivering
what we "want" from a couple basic search terms. How efficient is it
to type in a couple words and get a few million hits from which to
choose? I consider myself a fairly savvy search-nik, using all of the
advanced features to filter down to pages of interest. But when I get
down to the final result set it's still obvious to me which pages
aren't remotely related to what I really want. What's lacking is
context, the ability to separate text-oriented pages that include a
sequence of ASCII characters from data sources that have information
about a topic of interest. These are vastly different concepts and
obviously a cut above typing a word and clicking "I feel lucky".
I have no idea what ChunkIt does, but if it's a step closer to the
"semantic web" than what's available today then I will be very
interested in their progress.
(http://en.wikipedia.org/wiki/Semantic_Web)
I only hope they get/have some marketing savvy to explain the product
well and position it properly for its intended audience. This is an
area where they have failed dismally in the past - and no product will
survive poor marketing no matter how great it is.
>Also, they were very clear that customers who use their Pick products
>were not in their target audience for the TigerLogic product line.
I don't think that's a line to discourage a Pick shop from using their
other products, but to discourage a Pick shop from thinking this is
yet another Pick product that has an MD, BASIC, and multivalues. If a
site truly has a need for whatever a company offers I don't think
they'd discourage the enquiry.
>So, again, it seems they are a company wanting to leave their current
>customer base behind, except to the extent that it will pay annually
>for their non-strategic product line.
Well, in some ways I can't blame them.
- They add features that people don't use.
- People petition them to add features that can easily be coded in the
field.
- People complain about functionality that hasn't changed or bothered
anyone for a decade, yet somehow all of a sudden it's an industry
shaking show stopper.
- They can't approach SOA like IBM does for the U2 community because
most of the D3 user base wouldn't have a clue what they're talking
about.
People hang with D3 because it's functionally stable for their
purposes. Many sites only came to D3 because they were dragged from
AP and R83 before that. It's a market that has been pulled to the
current release, compared to other MV DBMS companies where the user
base and market requirements tend to drive development. In many ways
D3 engineers aren't any more progressive than their user base - it's
sort of a match. It's not an inherent limitation of the engineers
themselves or of D3 users either. I just think people in general tend
to look at D3 and just don't want or expect much change, so we don't
get it. There is a major disconnect between market demand and
improved revenue that results from product investment. Over time this
is perceived as product and even stagnation. So again, in some ways I
can't blame them.
>It looks like there have been
>enough changes that I should update that MV Family tree at some point,
>but it won't be this year. cheers! --dawn
So far, I think as far as MV goes the only thing that has changed the
name in our vendor files. I don't think the TigerLogic XDMS or
ChunkIt belong in the MV tree any more than Cognos, or DB2 - these are
peripheral offerings by companies that happen to also offer MV
products.
T
|
