Re: Market Share - flame

Alan, This is not directed at you. I fully understand the concept of purchasing based on 'market share', it is this that I find silly and flame-worthy. 1 - As I understand it the data on Market sharing came from two different sources (one for licenses, another for revenue). It is therefore not meaningful to derive a price factor from this information. 2 - Quoting Alan, quoting me. > Conservative companies and purchasing agents ("Nobody ever got fired for ^^^^^^^^^^^^^^^^^^^^^^^^^ > buying IBM.") like to buy from one of the [pick a number] top vendors in a ^^^^^^^^^^ No, but at times they should have been. > field. This assumes that the "market leaders" are less likely to go out of > business, leaving users holding the bag. > > ->It sounds to me like another Meaningless Indicator of Perfomance generated > ->by marketing types to convince unwary buyers that their product is better > ->than anybody else's. > > This is largely, but not entirely, true. What a silly way to do business! I haven't looked in Computerworld or Datapro, but we haven't seen data on this net that is not suspect (the license share data came from Informix). How can you look at data like that and say which is the top selling product? This is basically like trying to measure the output of a volcano with a pint cup. IT DOESN'T MATTER. If you are under the volcano you're pretty much dead anyway. On top of which the pint cup is going to melt as soon as you pick up your first scoop. There are basically 4 market leaders in the Unix RDBMS market, (Note, I'm not saying anything about OTHER operating systems or OOP databases), these are basically Informix, Oracle, Ingres and Sybase. Any of these product families will probably do fine for whatever your needs are. Ok, maybe product A won't do B, but then product C probably won't do X, Y or Z either. The best you can hope for is a product which allows you to develop work- arounds as necessary. All four of these PRODUCT families are likely to be around for a while. I know very little (I know I should stop there), about the specific companies. On top of which the concept of measuring market share in terms of percentage points is bogus. Each of these companies sells basically the same thing (an RDBMS) - but they each sell something completely different - Oracle has a wide span of platforms over which it runs, Sybase made their rep based on networking, Informix made their rep on a working product... Not too long ago I put out an RFI for a client for a Unix platform. The responses were amazing. Four of the companies were the market leaders! It turns out that one led the market in RISC platforms, a second let the market in a different area and so forth. In one sense the claims were completely bogus, but in truth each of these companies had a valid claim to market leadership in their own niche. So the thought of taking anybodys' data on market share at face value is somewhat amusing. Something like comparing the price of apples, oranges and bannanas... Would you buy a Ford Taurus just because it was the 'top selling car' last year? What if you want to transport 48 people at a time? Does a Ford Taurus still make sense? Market share is a similar statistic - It may be meaningful if ALL of the parameters are understood - what are we measuring, who is measuring, is this the product we are looking for or just a relative? Otherwise it's just hype. Yes I understand that market share is just another piece in the grand puzzle determining what product to purchase - and that nobody in their right mind (I hope) would dream of basing their purchase decision based solely on this piece of the puzzle. I just feel that it is such a SMALL piece of the puzzle as to be practically meaningless - unless of course you're talking about purchasing a product with .0023% of the market. j. _____________________________________________________________________________ Jack Parker - Contractor | Hewlett Packard, BSMC Boise, Idaho, USA| If you keep staring at it like that, jparker@hpbs2561.boi.hp.com | your nose is going to grow (208) 396-5388 (W) (208) 384-1623 (H) | into the bark. _____________________________________________________________________________