Re: IDS Feature Request List (including potential new requests).

rkusenet wrote: > "Two Pence" <twopence@peanutgallery.org> wrote in message > news:gPgzf.1790$AP6.1128@fe70.usenetserver.com... > >>How much effort _was_ it ( or _is_ it ) really? >> >>Isn't everything going into DB2 that comes from Informix? >> >>What better way to production test features without impacting >>the real target of customers with 'live development' on a smaller >>niche segment that is guaranteed not to grow much if at all. As >>long as you have that niche of customers willing and able to test it >>for you, why not! You did hear that feature development has the >>requirement to be doable in DB2 as well as Informix didn't you? > > > what do you mean? > > In the nearly 5 yrs after IBM acquired DB2, I still don't see DB2 having > Informix FRAGMENTATION. The worse technology term ever used in Informix was FRAGMENTATION, it should have been called partitioning to be consistent with it's competition calling it partitioning. DB2 partitioning is basically the same as Informix, and recently brought up a notch with range partitioning and a few other additions. DB2 also has MDC indexes, something I can't imagine for Informix given that it would probably require some features from XPS put into IDS--but I'm totally unqualified to even suggest that other than as a wild guess. DB2 has all the flexibility of XPS, if not more, in an OLTP engine. I still don't see DB2 having simple > ALTER TABLE DROP COLUMN :-)> I don't know, I thought this was added recently? > OTOH, Informix has DB2 style multiple buffer size, Oracle like online > Index, huge improvement in chunk/database size limit. Why is IBM's > priority screwed up? Shouldn't they be putting Informix features > in DB2, rather the other way. > I was going to ask you if you noticed BUFFERPOOLs in Informix 10, and LRU black magic and sorcery a thing of the past. DB2 still has better bufferpool management today, and a better instance architecture than Informix. The children of today can run DB2 with its high ease of use--easier to use than Informix, along with solid career incentives that don't exist with Informix. To be sure, HA replication is a great thing in Informix, but who cares? Only those that know about it, and quite honestly DB2 HADR is extremely easy to set up and manage. > And I don't necessarily see Informix morphing into Db2 as end of > Informix. Of course not it will simply EOL, and disappear. > If IBM can offer smooth migration free of cost and DB2 > has all features of Informix (even with totally different architecture), > then all fine for Informix customers. No Informix user will bemoan > loss of onstat etc, as long as he is getting something equal in DB2. > DB2 has a small set of annoyances, especially for Informix users. But in the grand scheme of things who really cares. DB2 will be compared more ( and is ) compared with SQL-Server or Oracle. This is the intended customer base, so all those little features that Informix has today that may or may not be in DB2 can easily be added to DB2 if necessary, but DB2 today is very good on its own merits in a non-Informix universe. The standards for performance in many large corporate environments shift with each passing succession of technology experts, and today's breed is not going to take Informix seriously unless there was some special feature that does not exist in DB2 that is a must have. It will take an existing, installed base of bitter- ender Informix die hards to keep it but you will only see isolated cases where Informix will ever be taken seriously again. Those days are over. Look around you and ask yourself if you can see the kids of today learning Informix enough to care about it 5 or 10 years from now. > Now if IBM had decided that Informix 9.4 was the last version (like they > said for XPS), then I agree with you. > > Like I said, $100 Million USD aint chump change, you and I would both love to have a piece of that, but as far as IBM is concerned, it's just a dollar amount contributing to the bottom line. No love lost ever, just elimination of non-performing products. And that's not to say it's the fault of Informix users, it's just the way the business has fallen into place, with all the right/wrong business decisions by former caretakers in a sea of products all clamoring for customers, developers and end-users.

Two Pence wrote: > rkusenet wrote: > >> "Two Pence" <twopence@peanutgallery.org> wrote in message ... >>> Isn't everything going into DB2 that comes from Informix? >>> >>> What better way to production test features without impacting >>> the real target of customers with 'live development' on a smaller >>> niche segment that is guaranteed not to grow much if at all. As >>> long as you have that niche of customers willing and able to test it >>> for you, why not! You did hear that feature development has the >>> requirement to be doable in DB2 as well as Informix didn't you? ... Hard to figure out who's friend or foe here ;-) Just a couple of clarifications: IDS is NOT the testbed for DB2 for LUW. You can take this as coming from the horse's mouth since I'm participating in DB2 feature planning. At no point point did anyone say: "Well.. that is risky, let's dump it on IDS first". ;-) The opportunity to compare two products without a marketing department in the middle spreading FUD allows for pretty honest soul searching and of course there is cherry picking both ways. HADR and "fragmentation" (aka range partitioning) being probably the two main attractions so far coming to DB2. Not every feature being added means technology transfer though. I'm sure the IDS developer who implemented SEQUENCEs in IDS did not need help from DB2 (or Oracle) dev any more than DB2 Dev needed help to implement DROP COLUMN. That's just industry convergence at work. W.r.t. to synching up new features, that's of course desirable. If one takes a look at the DB2 family it is easy to see that synching externals does not prevent innovation at any given product. E.g. DB2 for LUW was free to explore MQTs (aka. materialized views) years ahead of DB2 for zOS. The result of such synchronization is typically (at least on the SQL side) a much better, thought through externals design. Cheers Serge -- Serge Rielau DB2 Solutions Development DB2 UDB for Linux, Unix, Windows IBM Toronto Lab

Serge Rielau said: > > E.g. DB2 for LUW was free to explore MQTs (aka. materialized views) > years ahead of DB2 for zOS. But I thought they had the same code base? -- Bye now, Obnoxio "C'est pas parce qu'on n'a rien ' dire qu'il faut fermer sa gueule" - Coluche did i mention i like nulls? heck, i even go so far as to say that all columns in a table except the primary key could/should be nullable. this has certain advantages, for example, if you need to insert a child record and you don't have a parent row for it, just do an insert into the parent table with the primary key value (everything else null), and voila, relational integrity is preserved. but this is, admittedly, a bit controversial among modellers. --r937, dbforums.com

Serge Rielau wrote: > Two Pence wrote: > >> rkusenet wrote: >> >>> "Two Pence" <twopence@peanutgallery.org> wrote in message > > ... > >>>> Isn't everything going into DB2 that comes from Informix? >>>> >>>> What better way to production test features without impacting >>>> the real target of customers with 'live development' on a smaller >>>> niche segment that is guaranteed not to grow much if at all. As >>>> long as you have that niche of customers willing and able to test it >>>> for you, why not! You did hear that feature development has the >>>> requirement to be doable in DB2 as well as Informix didn't you? > > ... > Hard to figure out who's friend or foe here ;-) > > Just a couple of clarifications: > IDS is NOT the testbed for DB2 for LUW. You can take this as coming from > the horse's mouth since I'm participating in DB2 feature planning. > At no point point did anyone say: > "Well.. that is risky, let's dump it on IDS first". ;-) > You may not have said that verbally, but it does follow that Informix is the step-child, or better, the rescued dead-man-walking, it will always take a back seat, second fiddle to DB2. If you think about the acquisition of a company, and merging the acquired company assets, certainly there would have to be limited/controlled exposure to the parent company of those newly acquired assets as well as making sure that you don't allow the newly acquired company to have any market momentum that cannot be lock-stepped with the parent. Informix is not going to go off on its own and become greater than DB2, this simply cannot happen. It may be fine and dandy down at the functional level to suggest that Informix is not dead, but in reality it _is_ dead to the rest of the market, having been spared a death sentence, only now living under the mercy and cover of its savior. It looks that way because it IS that way. It needs a new identity, and a new owner, so that it can move on and evolve. Anything you say that says Informix has free reign to make its destiny great sounds like a lie simply because of how people see DB2 and Informix in the market. Informix 10 looks like a great product but without REAL marketing it stands no chance in the market. It looks like a step child because it IS one. It might be good for Informix to be spun off on its own, and allowed to compete as much as it wants to, yes, even with its savior. As long as it identifies with being DB2's family sibling, it will forever languish in limbo. It will be entombed by being embedded. Interestingly enough MySQL finds great pride in describing itself as a leader in the embedded space, maybe this is where IBM wants to position Informix. But that shows a great lack of understanding of the Informix product line. > The opportunity to compare two products without a marketing department > in the middle spreading FUD allows for pretty honest soul searching and > of course there is cherry picking both ways. > HADR and "fragmentation" (aka range partitioning) being probably the two > main attractions so far coming to DB2. I remain curious how you can market two peer products owned by the same company without confusing the customer. It would be the same for a car manufacturer or any other line of business. People see right through it. > Not every feature being added means technology transfer though. I'm sure > the IDS developer who implemented SEQUENCEs in IDS did not need help > from DB2 (or Oracle) dev any more than DB2 Dev needed help to implement > DROP COLUMN. > That's just industry convergence at work. > > W.r.t. to synching up new features, that's of course desirable. > If one takes a look at the DB2 family it is easy to see that synching > externals does not prevent innovation at any given product. > E.g. DB2 for LUW was free to explore MQTs (aka. materialized views) > years ahead of DB2 for zOS. > The result of such synchronization is typically (at least on the SQL > side) a much better, thought through externals design. > > Cheers > Serge

This is like saying there is no reason for Audi to exist because because VW has its own products so it should just shut down Audi. Some of their cars even share the same platform. From a marketing perspective it is completely acceptable for IBM to have both DB2 and Informix. They can position them for diffrent markets. They can roll different features into them to better serve their respective markets. They could guild one and sell it as a premium product and the other as the entry level value product. They can share code to make each product better. I see nothing but synergy. Even where they compete against each other its a fight that IBM wins either way. If Informix has customer loyalty (which it does) and is making a profit for IBM then it is a winning proposition to keep Informix alive as a product. If DB2 occasionaly steals a feature from Informix it is just another way for IBM to leverage its ownership of Informix. If Informix steals a feature from DB2 again it is just away to get more value out of their development dollars. Porting a feature from DB2 to Informix is probably cheaper and safer than starting a feature from scratch in Informix and vice-versa. Even if both products converge to having the same code base they could still position the products differently. For example, DB2 Express which shares 90% of its code base with DB2 may even become a free product to prime the pump for companies to use DB2 for bigger projects. Here is a nice article saying the exact opposite about Informix and DB2. http://www.regdeveloper.co.uk/2006/01/18/db2_neglected/ So in summary FOD. ;-)

bozon wrote: > This is like saying there is no reason for Audi to exist because > because VW has its own products so it should just shut down Audi. Some > of their cars even share the same platform. I have failed to make the car analogy communicate my point, and for that you have my apologies. VW owns Audi and VW products. But if VW owned BMW and Audi, this would be confusing to the customer. This is where Informix and DB2 can make the analogy. They are peer products and it is glaringly obvious to outsiders, and this is the point. If IBM didn't see this, then they wouldn't have any problem screaming the Informix name out there in a marketing campaign like they do with DB2. > From a marketing > perspective it is completely acceptable for IBM to have both DB2 and > Informix. You are a spokesperson for IBM? I wasn't aware of a marketing program for either product, other than an occasional ad in print magazines for DB2, and a few token mentions of Informix. When there _is_ a DB2 ad, one hardly sees Informix, and for all the reasons I mentioned. > They can position them for diffrent markets. They can roll > different features into them to better serve their respective markets. Pigs will soon fly I'm confident of that. > They could guild one and sell it as a premium product and the other as > the entry level value product. I'm sure Informix customers and those who want to promote it will feel great about Informix being "an entry level product" . > They can share code to make each product > better. I see nothing but synergy. Even where they compete against each > other its a fight that IBM wins either way. You're getting the good stuff to smoke I'm sure of it. > If Informix has customer > loyalty (which it does) and is making a profit for IBM then it is a > winning proposition to keep Informix alive as a product. It is interesting that you wouldn't consider Informix outside its base of existing customers. One would have to consider those customers migrating away from Informix to whatever other product is out there if you really were thinking about the market and those customers objectively. > If DB2 > occasionaly steals a feature from Informix it is just another way for > IBM to leverage its ownership of Informix. If Informix steals a feature > from DB2 again it is just away to get more value out of their > development dollars. Porting a feature from DB2 to Informix is probably > cheaper and safer than starting a feature from scratch in Informix and > vice-versa. > Oh ok. > Even if both products converge to having the same code base they could > still position the products differently. For example, DB2 Express > which shares 90% of its code base with DB2 may even become a free > product to prime the pump for companies to use DB2 for bigger > projects. > Here is a nice article saying the exact opposite about > Informix and DB2. The article had a lot to say about DB2 but I didn't see but one sentence about Informix. Maybe you got a different version of the website than I did. > http://www.regdeveloper.co.uk/2006/01/18/db2_neglected/ > > So in summary FOD. ;-) > I thought your name was bozon?

>I have failed to make the car analogy communicate my point, and for that >you have my apologies. VW owns Audi and VW products. But if VW owned >BMW and Audi, this would be confusing to the customer. This is where >Informix and DB2 can make the analogy. They are peer products and it >is glaringly obvious to outsiders, and this is the point. Ford - Volvo, Lincoln - Jaguar, many companies have "peer" products and manage to diffrentiate them to the marketplace. Coke, sprite, barq's all are beverages aimed at different market segments. > > From a marketing >> perspective it is completely acceptable for IBM to have both DB2 and >> Informix. >You are a spokesperson for IBM? No, and it isn't important for me to have my opinion on the matter. > I wasn't aware of a marketing program >for either product, other than an occasional ad in print magazines for >DB2, and a few token mentions of Informix. When there _is_ a DB2 ad, >one hardly sees Informix, and for all the reasons I mentioned. Advertising is not marketing. Marketing is positioning a product and creating a strategy to sell a product. If you never see an ad for a product it doesn't mean that it isn't being marketed. >The article had a lot to say about DB2 but I didn't see but one sentence about Informix. Maybe you got a different version of the website than I did. "So, how come you don't use it? Let's face it, when most people argue about databases, it's the same old Oracle vs. SQL Server fight, with maybe some Sybase and Informix thrown in. Who mentions DB2? Yet we should." Which is the one sentence that I thought summarized the point of the article which is DB2 gets less respect in the marketplace than any other database including Informix. >Pigs will soon fly I'm confident of that. Translation: I can't make a reasoned arguement so I'll make a bad joke. >> If Informix has customer >> loyalty (which it does) and is making a profit for IBM then it is a >> winning proposition to keep Informix alive as a product. >It is interesting that you wouldn't consider Informix outside its >base of existing customers. One would have to consider those >customers migrating away from Informix to whatever other product >is out there if you really were thinking about the market and those >customers objectively. I considered this as a bare minimum requirement. I believe that Informix is doing much better than keeping its existing customers. >> better. I see nothing but synergy. Even where they compete against each >> other its a fight that IBM wins either way. >You're getting the good stuff to smoke I'm sure of it. Translation: I can't make a reasoned arguement so I'll make a bad joke. >> So in summary FOD. ;-) You'll have to search the web for this. >I thought your name was bozon? Bozon - It is the elemental particle of stupidity of which all other stupidity is composed. This particle is indivisible and cannot be broken down into any smaller parts of stupidity.

bozon wrote: > Ford - Volvo, Lincoln - Jaguar, many companies have "peer" products and > manage to diffrentiate them to the marketplace. Coke, sprite, barq's > all are beverages aimed at different market segments. > OK, so how do you propose, in a market of what, 3 or 4 major databases, to differentiate Informix from DB2? With those differences, why would you pick Informix instead of DB2? What is the motivator??? What catchy marketing line would you use to describe Informix?

bozon wrote: > This is like saying there is no reason for Audi to exist because > because VW has its own products so it should just shut down Audi. Some > of their cars even share the same platform. From a marketing > perspective it is completely acceptable for IBM to have both DB2 and > Informix. I agree but the marketplace doesn't. It is not just IBM that matters. Without active support from SAP, PeopleSoft, eBusiness Suite, JD Edwards, Baan, Siebel any database will be marginalized. So try your car analogy when no one wants to sell the Audi owner petrol. How long do you think the Audi would last except as a curiosity? But to continue the discussion. How many word processors does Microsoft sell? How many spreadsheets? But you might argue that they have three different databases and you would be correct. SQL Server, Access, and if your memory is really good Fox. Remember Fox? Oracle has RDB? Remember RDB? And in the same light IBM has Informix. SQL Server and Access can co-exist because there is no question which one is the real database and which one is a toy. When I see Informix links on DB2 pages like I see DB2 links on virtually every Informix page ... I will start to believe they are equals. When I see developers of major commercial applications supporting Informix I will truly believe and likely so will the marketplace. -- Daniel A. Morgan http://www.psoug.org damorgan@x.washington.edu (replace x with u to respond)

bozon wrote: > Ford - Volvo, Lincoln - Jaguar, many companies have "peer" products and > manage to diffrentiate them to the marketplace. Coke, sprite, barq's > all are beverages aimed at different market segments. You are correct. But the problem with differentiating between DB2 and Informix is that most people know Informix is technically the superior product and they just don't buy IBM's positioning. -- Daniel A. Morgan http://www.psoug.org damorgan@x.washington.edu (replace x with u to respond)

DA Morgan wrote: > bozon wrote: > >> Ford - Volvo, Lincoln - Jaguar, many companies have "peer" products and >> manage to diffrentiate them to the marketplace. Coke, sprite, barq's >> all are beverages aimed at different market segments. > > > You are correct. But the problem with differentiating between DB2 and > Informix is that most people know Informix is technically the superior > product and they just don't buy IBM's positioning. How is Informix 'technically the superior product' to DB2? I would really like to know. But I would be willing to bet you nor anybody here will be able to say. Informix is not 'technically superior' to DB2 and that's the reality of it all. But technical superiority has little to do with the success of a database product, as an Oracle person you should be able to speak first hand about that. :-)

Two Pence wrote: > DA Morgan wrote: > >> bozon wrote: >> >>> Ford - Volvo, Lincoln - Jaguar, many companies have "peer" products and >>> manage to diffrentiate them to the marketplace. Coke, sprite, barq's >>> all are beverages aimed at different market segments. >> >> >> >> You are correct. But the problem with differentiating between DB2 and >> Informix is that most people know Informix is technically the superior >> product and they just don't buy IBM's positioning. > > > > How is Informix 'technically the superior product' to DB2? > > I would really like to know. > > But I would be willing to bet you nor anybody here will be able to say. > Informix is not 'technically superior' to DB2 and that's the reality of > it all. But technical superiority has little to do with the success of a > database product, as an Oracle person you should be able to speak first > hand about that. :-) Not an Oracle person ... a Fortran to Cobol to DB2 to Oracle person. My experience with Informix, granted it was some years ago, left a definite impression that ... given the choice ... I'd rather be in Informix than DB2 were that the choices on the table. But then I'm one of those people that likes Sybase more than SQL Server and we all know where that one is going to end up too. -- Daniel A. Morgan http://www.psoug.org damorgan@x.washington.edu (replace x with u to respond)

bozon said: > >> someone else that bozon didn't name said: >> >>I have failed to make the car analogy communicate my point, and for that >>you have my apologies. VW owns Audi and VW products. But if VW owned >>BMW and Audi, this would be confusing to the customer. This is where >>Informix and DB2 can make the analogy. They are peer products and it >>is glaringly obvious to outsiders, and this is the point. > > Ford - Volvo, Lincoln - Jaguar, many companies have "peer" products and > manage to diffrentiate them to the marketplace. Coke, sprite, barq's > all are beverages aimed at different market segments. Why the fuck is it any different if VW own Audi and VW or if VW owns Audi and BMW? And anyway, VW owns VW, Audi, Seat, Skoda, Bentley and Lamborghini. Now the latter two are easy enough to split from the mix because of their niche nature, but there is an incredible amount of overlap between Skoda -> Seat -> VW -> Audi. VW have just figured out how to position these brands in different market. -- Bye now, Obnoxio "C'est pas parce qu'on n'a rien ' dire qu'il faut fermer sa gueule" - Coluche did i mention i like nulls? heck, i even go so far as to say that all columns in a table except the primary key could/should be nullable. this has certain advantages, for example, if you need to insert a child record and you don't have a parent row for it, just do an insert into the parent table with the primary key value (everything else null), and voila, relational integrity is preserved. but this is, admittedly, a bit controversial among modellers. --r937, dbforums.com

Two Pence said: > > bozon wrote: >> Ford - Volvo, Lincoln - Jaguar, many companies have "peer" products and >> manage to diffrentiate them to the marketplace. Coke, sprite, barq's >> all are beverages aimed at different market segments. >> > > OK, so how do you propose, in a market of what, 3 or 4 major databases, > to differentiate Informix from DB2? With those differences, why would > you pick Informix instead of DB2? What is the motivator??? What catchy > marketing line would you use to describe Informix? I daresay it's not Mr Bozon's job to do that. But given the VW / Skoda / Seat / Audi analogy, it's definitely possible. -- Bye now, Obnoxio "C'est pas parce qu'on n'a rien ' dire qu'il faut fermer sa gueule" - Coluche did i mention i like nulls? heck, i even go so far as to say that all columns in a table except the primary key could/should be nullable. this has certain advantages, for example, if you need to insert a child record and you don't have a parent row for it, just do an insert into the parent table with the primary key value (everything else null), and voila, relational integrity is preserved. but this is, admittedly, a bit controversial among modellers. --r937, dbforums.com

Two Pence said: > > You may not have said that verbally, but it does follow that Informix is > the > step-child, or better, the rescued dead-man-walking, it will always take > a back seat, second fiddle to DB2. If you think about the acquisition of > a company, and merging the acquired company assets, certainly there would > have > to be limited/controlled exposure to the parent company of those newly > acquired > assets as well as making sure that you don't allow the newly acquired > company > to have any market momentum that cannot be lock-stepped with the parent. > Informix > is not going to go off on its own and become greater than DB2, this simply > cannot > happen. > > It may be fine and dandy down at the functional level to suggest that > Informix is > not dead, but in reality it _is_ dead to the rest of the market, having > been spared > a death sentence, only now living under the mercy and cover of its savior. > It looks > that way because it IS that way. It needs a new identity, and a new > owner, so that > it can move on and evolve. Anything you say that says Informix has free > reign to make > its destiny great sounds like a lie simply because of how people see DB2 > and Informix > in the market. Informix 10 looks like a great product but without REAL > marketing it > stands no chance in the market. It looks like a step child because it IS > one. > > It might be good for Informix to be spun off on its own, and allowed to > compete as > much as it wants to, yes, even with its savior. As long as it identifies > with > being DB2's family sibling, it will forever languish in limbo. It will be > entombed > by being embedded. Interestingly enough MySQL finds great pride in > describing itself > as a leader in the embedded space, maybe this is where IBM wants to > position Informix. > But that shows a great lack of understanding of the Informix product line. > > I remain curious how you can market two peer products owned by the same > company without confusing the customer. It would be the same for a car > manufacturer or any other line of business. People see right through it. Tim, how many fucking email addresses do you have, anyway? -- Bye now, Obnoxio "C'est pas parce qu'on n'a rien ' dire qu'il faut fermer sa gueule" - Coluche did i mention i like nulls? heck, i even go so far as to say that all columns in a table except the primary key could/should be nullable. this has certain advantages, for example, if you need to insert a child record and you don't have a parent row for it, just do an insert into the parent table with the primary key value (everything else null), and voila, relational integrity is preserved. but this is, admittedly, a bit controversial among modellers. --r937, dbforums.com

Two Pence said: > > I have failed to make the car analogy communicate my point, and for that > you have my apologies. VW owns Audi and VW products. But if VW owned > BMW and Audi, this would be confusing to the customer. This is where > Informix and DB2 can make the analogy. They are peer products and it > is glaringly obvious to outsiders, and this is the point. If IBM didn't > see this, then they wouldn't have any problem screaming the Informix > name out there in a marketing campaign like they do with DB2. Ah! It was you. I might have guessed. I hate to piss in your chips, but as I pointed out in my other post, there is a MASSIVE overlap between very similar cars, were the branding of the products has been carefully defined by VW AG. They also took a famously shite brand (Skoda) and rebranded it as the cost-concious quality car, so they're more austere than VW's but just as well built. Seat's are branded as sporty, VW as quality, reliable cars and Audi as the upmarket car. At the end of the day, they're all made to pretty much the same standard and feel equally good to drive, it's purely down to how the cars are marketed and the enforced differences between them that make markets for all of them. While there is cannibalism, the consequence of all these options is that their total market share is doing well, better than if they just had the VW brand to play with. Of course, this DOES all require an organisation that can actually market stuff. -- Bye now, Obnoxio "C'est pas parce qu'on n'a rien ' dire qu'il faut fermer sa gueule" - Coluche did i mention i like nulls? heck, i even go so far as to say that all columns in a table except the primary key could/should be nullable. this has certain advantages, for example, if you need to insert a child record and you don't have a parent row for it, just do an insert into the parent table with the primary key value (everything else null), and voila, relational integrity is preserved. but this is, admittedly, a bit controversial among modellers. --r937, dbforums.com

Obnoxio The Clown wrote: > Serge Rielau said: > >>E.g. DB2 for LUW was free to explore MQTs (aka. materialized views) >>years ahead of DB2 for zOS. > > > But I thought they had the same code base? > Only when you run statistics. -- Serge Rielau DB2 Solutions Development DB2 UDB for Linux, Unix, Windows IBM Toronto Lab

DA Morgan wrote: > Not an Oracle person ... a Fortran to Cobol to DB2 to Oracle person. > My experience with Informix, granted it was some years ago, left a > definite impression that ... given the choice ... I'd rather be in > Informix than DB2 were that the choices on the table. But then I'm > one of those people that likes Sybase more than SQL Server and we all > know where that one is going to end up too. Hmm, did'you rile just a few posts up about the need for continued education? How can you make statements about Informix and DB2 if you know only "neolithic" versions of one or both? Cheers Serge -- Serge Rielau DB2 Solutions Development DB2 UDB for Linux, Unix, Windows IBM Toronto Lab

Obnoxio The Clown wrote: > Two Pence said: > >>I have failed to make the car analogy communicate my point, and for that >>you have my apologies. VW owns Audi and VW products. But if VW owned >>BMW and Audi, this would be confusing to the customer. This is where >>Informix and DB2 can make the analogy. They are peer products and it >>is glaringly obvious to outsiders, and this is the point. If IBM didn't >>see this, then they wouldn't have any problem screaming the Informix >>name out there in a marketing campaign like they do with DB2. > > > Ah! It was you. I might have guessed. I hate to piss in your chips, but as > I pointed out in my other post, there is a MASSIVE overlap between very > similar cars, were the branding of the products has been carefully defined > by VW AG. > Don't most car organizations go out of their way to differentiate themselves, even to the point of having different CEOs, dealerships, media campaigns, etc? I see this a lot with many of the car brands out there. So, how will you do that with Informix and DB2 under the current scheme? Who gets to be the bitch? These two products are under the same organization, the same "branding" operation--or did I miss the last round of Information Management naming thing recently where they drop the DB2 part of it. If you're going to play VW, then I expect separate dealerships, one for Audi and one for the Jetta. Is this going to happen for IBM's line-up? I doubt it. They don't even understand what the Informix product line consists of, or what the various products are. And just exactly what will be the differentiating thing between them? One works well as an embedded database, the other as the bright and shining super db. One represents IBM, the other represents the people that get sealed into the pyramid now that the king has died, the 'legacy' db from back-in-the-day, good-bye, see you on the other side. > They also took a famously shite brand (Skoda) and rebranded it as the > cost-concious quality car, so they're more austere than VW's but just as > well built. Seat's are branded as sporty, VW as quality, reliable cars and > Audi as the upmarket car. At the end of the day, they're all made to > pretty much the same standard and feel equally good to drive, it's purely > down to how the cars are marketed and the enforced differences between > them that make markets for all of them. > So when do we see MySQL vs Standard Engine? This would be consistent with what you're saying. It's about the only place you can throw an Informix product out there that DB2 really can't go up against in the ISAM space. > While there is cannibalism, the consequence of all these options is that > their total market share is doing well, better than if they just had the > VW brand to play with. > > Of course, this DOES all require an organisation that can actually market > stuff. > One could only hope. And your theory also suggests separate sales organizations of some kind, rather than one sales crew having to pitch two separate product lines.

Two Pence wrote: > > How is Informix 'technically the superior product' to DB2? > > I would really like to know. > Currently ? Signficantly stronger partitioning and indexing capabilities Better online operation/schema evolution capabilities Less funky size limitations HADR for Informix has less restrictions than the DB2 equivalent Most of the utilities - onwhatevers Smarter users ? I still think DB2 has aways to got to catch Informix in some areas. In other areas (SQL support, marketing) they are ahead.

Two Pence wrote: > What catchy marketing line would you use to describe Informix? Damn the torpdeos, full speed ahead ? (Oh wait, that one's been done)

DA Morgan Said >But to continue the discussion. How many word processors does Microsoft sell? How many spreadsheets? At least 2 word processors and 2 spreadsheets, because you can't forget about Microsoft Works. Of course I wouldn't want to be the "Microsoft Works" of databases. They also sell several different opperating systems. XP Home, Pro, That damn PDA operating system that I can't think of. I was originally a Sybase person and couldn't believe when I had to do an Oracle project that Oracle (Oracle 6) didn't have a cost based optimizer. I thought what is the point if I have to tell the server how to execute the query, but the savings in development costs and getting to market quicker made up for the lack of features. They could apply the savings on development to salespeople. "Sure its shit now but wait til version 7." I think was their marketing campaign.