Re: Does anyone read the e-mails from [mailto:db2@news.db2mag.com]
Not a technical support thread — it's a marketing/strategy debate about IBM's handling of Informix IDS. Participants argue over whether published TPC benchmarks and a developer-level "grass roots" push are what IDS needs to regain visibility against Oracle and DB2, with criticism of IBM Information Management management for ignoring advice. Mark Townsend corrects the claim that Oracle no longer publishes benchmarks (citing recent TPC-C/TPC-H results), which Gumby concedes, then asks where IDS's benchmarks are. The thread drifts into jokes about blogs and Drizzle; no technical problem or resolution is recorded.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
On Aug 1, 9:45 pm, InDeep <ind...@indeep.com> wrote: > Gumpy, > > You are not quite fair to IBM. I went out to their DB2mag website and > I see Carltons' book out there on the front page. IBM is not exactly ignoring > IDS. But I get your points. Without the benchmark, IDS will languish in the > cloud/shadow of unknownness (<--is that a word? :-) among the mainstream > community of customers and developers. > I'm very fair to IBM. I mean there's a lot I could say that I don't. I pick on issues that at the surface don't cause one to ask too many questions, most of the time.;-) Its not so much the benchmark in and of itself. And I have to explain this... A benchmark is a talking point. Choose the right combination of hardware, software and configuration, you have a benchmark that will last a while. Depending on the product choices and overall costs, your benchmark may last 18 months, or until your competitor decides to outspend you on hardware. With benchmarking, you have two races. Ultimate max power, and max efficiency. Lowest cost per transaction. But its the talking point you want to make. You want to give your sales reps, your marketing people, the "press" something to talk about. Oracle doesn't. And there's a reason for this. In IBM sales speak, Oracle is in column A. All the rest are column fodder. If you're in column A, you want to dismiss any talking points that could jeopardized your being in column A. If you're in column B, C or D, then you want to raise talking points that shine the light on your product, and show the flaws of your competition to get you in to column A and then close the deal. Oracle doesn't compete in benchmarks anymore, or at leas we haven't seen any TPC-C or TPC-E benchmarks released lately. There's a reason for this. They don't have to do benchmarks because they've secured their spot at the 800 pound gorilla in column A. You want to displace them? You're going to need a lot of effort in different areas. Benchmarking is the simplest way to create a talking point. Oh there's more, but its not my job to teach a bunch of DBAs the art of sales. ;-) > You bring up a lot of points, and well-focused. But I feel that you are treating > the symptoms rather than the problem. It really boils down to getting the grass > roots thang going. Is that possible with the current options? Unknown. But > back to the point, it's really all about grass roots, all the other methods have > been sub-marginally tried and given their results: IDS is important but > not necessarily even a choice among the unwashed masses. Grass roots is a proven > formula, but it only exists today among the faithful. The benchmark issue is one that > really deserves more attention to build the faith. While there are many arguments > to the validity of benchmarks, it really is all about the marketing. Marketing to > not only the business community, and the current faithful, but also to the new > generation of the developer community, who need a real reason to choose IDS over > other options, not just to install it, but to really spend the time to develop > applications with IDS. This is really where the grass roots movement is important, > at the developer level ( read new developers ) . > First, my points are *not* well focused. They are for the most part, written in short blips of time when I need a distraction. Sure the ideas behind them aren't off the cuff. They're a culmination of several years of observation. I also don't say all that I can say. An example, I gave a very basic explanation of the sales process. You also act as if the grass roots is key. It again by itself is not. You can plant sod on clay soil, but by itself, even with water, the sod will not take. You talk about it working among its "faithful". Its not. Did you look at the IIUG e-mailer? (They met this weekend.) Gary Ben-Israel laments that each passing day it gets harder to explain his decisions to his superiors even though he knows he made the correct choice in staying with Informix/IDS. In addition. If you look at all of the blogs being written... Well, how many are on IBM? How many are on Informix fan sites? And how many are on mainstream press sites? If you said the last one was >0 I'd have to tell you that you're off base and need to redefine what you think is "mainstream". The grass roots campaign isn't one that feeds in to the existing base. Sure you need that to gain momentum. But beyond that? You've hit a wall. Its more than marketing. And that's a discussion we won't have here. That's one for the Ambush, Jerry K, Alyse Pascarelli ?Sp?, and a few others. But that conversation will never happen. Ever see an IBM senior exec eat crow? They don't. Its not in their nature to admit they fscked up. > Gumpy like you I've been there for the whole ride. I remember the Phil White > magazine ads, the heady times of watching 7x smoke the competition. But now its > all about the groundswell. Most of the stuff coming out now whether its open > source or other is all about the buzz at the developer level. Until _that_ group > is kicked in the ass about the dazzle surrounding the product, you, as well as > anyone else are academic about all of it. I not only keep hoping against the odds > that IDS will come alive in the market, I also want a pic of OTC on a unicycle > screaming through the streets "IDS! IDS! IDS!". Both are highly unlikely, but > at this point I'm more hopeful for the pic of the clown. :-) > The whole ride? LOL... You missed the first part of the ride. When Sippl stepped aside to bring in Phil White. And Phil made all the right moves. The only major real screw up was in how long it would take to get Illustra in to the engine. After that, it was all down hill. Not even their choices for a successor were good. You want the truth, IDS isn't dead and it stands a chance. But IBM's IM senior management just doesn't grok it. IBM's IM management have been told repeatedly on how to improve the image and marketshare of IDS. Yet they haven't done it. That's the question you have to ask yourself. Why haven't they? But hey! What do I know? ;-) Definitely more than I can say.
> Oracle doesn't compete in > benchmarks anymore, or at leas we haven't seen any TPC-C or TPC-E > benchmarks released lately. Stop making crap up. Oracle publishes TPC-C's and TPC-H's regulary. This year alone, Oracle has published 2 TPC-Cs, and a TPC-H. This is one more TPC-H than IBM has published, and 1 less TPC-C. Nobody other than MS is publishing TPC-E's at this stage.
On Aug 5, 3:12 am, Mark Townsend <markbtowns...@sbcglobal.net> wrote: > > Oracle doesn't compete in > > > benchmarks anymore, or at leas we haven't seen any TPC-C or TPC-E > > benchmarks released lately. > > Stop making crap up. Oracle publishes TPC-C's and TPC-H's regulary. This > year alone, Oracle has published 2 TPC-Cs, and a TPC-H. This is one more > TPC-H than IBM has published, and 1 less TPC-C. > > Nobody other than MS is publishing TPC-E's at this stage. Hmmm. I went back and looked. Yup you're right. There's Oracle. Two benchmarks. One on dell with the lowest cost per tpm, still beating out Dell and Sybase. But then there's a ho-hum benchmark where it topped out at a little over a million tpm but the cost was twice that of DB2 that did 6 million tpm. So mea culpa. Even the 800 pound gorilla does benchmarks to show their prowess. So where's IDS?
Ian Michael Gumby said: > > Go! Drizzle! (You have to read el reg to grok that !) Drivel? That's soooo yesterday, doncha know? http://obotheclown.blogspot.com/2008/08/back-into-darkened-room-i-go.html -- Bye now, Obnoxio http://obotheclown.blogspot.com/
Obnoxio The Clown wrote: > Ian Michael Gumby said: >> Go! Drizzle! (You have to read el reg to grok that !) > > Drivel? That's soooo yesterday, doncha know? > > http://obotheclown.blogspot.com/2008/08/back-into-darkened-room-i-go.html > From the photos, I reckon Brian Aker is OTC after a walk in the rain. Having one blog where you post provocative essays and another where you ridicule them under a nom-de-plume is a sure-fire way to drive traffic to both isn't it? ;-) Just my '0.02 worth. -- RGB How often should you UPDATE STATISTICS on Drizzle?
RedGrittyBrick said: > Obnoxio The Clown wrote: >> Ian Michael Gumby said: >>> Go! Drizzle! (You have to read el reg to grok that !) >> >> Drivel? That's soooo yesterday, doncha know? >> >> http://obotheclown.blogspot.com/2008/08/back-into-darkened-room-i-go.html > > From the photos, I reckon Brian Aker is OTC after a walk in the rain. Up yours, sideways! > Having one blog where you post provocative essays and another where you > ridicule them under a nom-de-plume is a sure-fire way to drive traffic > to both isn't it? ;-) Er, what? It hadn't occurred to me, but now that you mention it... :o> So, who is mocking me, or who am I mocking? :os -- Bye now, Obnoxio http://obotheclown.blogspot.com/