Re: These crazy emails about IDS to DB2 conversion

Malcolm W more or less hit the nail on the head. My input to this is, I'm an Informix DBA and I'd rather not have to learn how to DBA another vendor's offering. This statement may or may not be rational but hey, that's me, and I'm old. What I guess we all want to know is whether the impending merge of technology (as I recall, the working name quoted at the Infobahn was 'DBNew' but what's in a name?) will: A) retain the capabilities and performance of IDS B) be an upgrade which will be no more difficult than a version upgrade, i.e. take old server down; install new server; start new server and allow conversion to occur; go online. 1.5 hours, max. C) Retain the administration capabilities of IDS D) run our 4GL/SQL/ESQLC etc without mods. If any of these aren't met, will the transition at best be no more than a weekend's work to iron out? Further to that; are there any *really* good reasons we shouldn't look at Oracle, Sybase, SQLServer etc. if the transition to DBNew is more difficult than we'd like. And we don't need answers that begin "We all know that...." or "It's plainly obvious...." and so on. Does IBM have a hidden agenda to shaft Informix users? No speculation, please. Those of us who are NOT close to the front line, those of us who just DBA or write code every day, don't know the wheres, hows or whys of the cynicism, but it certainly makes the confidence in our chosen technology plummet. You guys haven't got a CLUE the effect all the bad-mouthing of IBM that goes on here is having on IT management teams - there are corporate decision-makers reading this group. Mine are saying "We've got to drop Informix like NOW before they force us up sh*t creek and take away our paddle". In as many words. I fear for my job. I'm beginning to feel I'd be better off busking. So, how do we rebuild that confidence? All I hear here is bad-mouthing and little that's constructive. 2.5 pence from Malc_P

Malc P wrote: > Malcolm W more or less hit the nail on the head. If he did, then why are you asking questions about IBM's IDS strategy? :o) > My input to this is, I'm an Informix DBA and I'd rather not have to > learn how to DBA another vendor's offering. This statement may or may > not be rational but hey, that's me, and I'm old. > What I guess we all want to know is whether the impending merge of > technology (as I recall, the working name quoted at the Infobahn was > 'DBNew' but what's in a name?) will: > > A) retain the capabilities and performance of IDS > B) be an upgrade which will be no more difficult than a version > upgrade, i.e. take old server down; install new server; start new > server and allow conversion to occur; go online. 1.5 hours, max. > C) Retain the administration capabilities of IDS > D) run our 4GL/SQL/ESQLC etc without mods. > > If any of these aren't met, will the transition at best be no more > than a weekend's work to iron out? Not if the administration doesn't get resolved. Conversion will almost certainly depend on data set size. I strongly doubt that DB2 will be able to match the more esoteric aspects of IDS without considerable work, it may be better to wrap a DB2 wrapper around IDS than try to implement all the Illustra APIs. There is talk that 4GL will be given DRDA (IBM data access) at some unspecified point in the future, as there is talk that 4GL will somehow be integrated into IBM's EGL strategy. I'm not aware of any specifics on either of these. Allegations have been made that some of the onstats may be put into a forthcoming version of DB2, but I'm not sure there will ever be a full merge of the administration between the two. (That's just an opinion, mind, I have no concrete reason for saying that.) > Further to that; are there any *really* good reasons we shouldn't look > at Oracle, Sybase, SQLServer etc. if the transition to DBNew is more > difficult than we'd like. We all know that it's plainly obvious that there isn't a really good reason not to look at these if DBNew doesn't cut the mustard. Price, quality, tech support, all the usual. > And we don't need answers that begin "We all know that...." or "It's > plainly obvious...." and so on. Oh, sorry. > Does IBM have a hidden agenda to shaft Informix users? No speculation, > please. Those of us who are NOT close to the front line, those of us > who just DBA or write code every day, don't know the wheres, hows or > whys of the cynicism, but it certainly makes the confidence in our > chosen technology plummet. IBM doesn't have any agenda at all, from my dealings with them. The whole reason I and several other dishonest and slovenly people are moaning is because the chosen technology is so much better than anything else and I, for one, do not feel that IBM is doing IDS justice, for want of a better term. But to be fair to IBM, this just seems to be the way they work. I've heard that someone has woken up to the fact that NUMA-Q was actually quite spiffy technology, and now they are scrambling around trying to find the Sequent people that they dispersed or lost after the acquisition. Or maybe they just need them for the fight with SCO. :o) > You guys haven't got a CLUE the effect all the bad-mouthing of IBM > that goes on here is having on IT management teams - there are > corporate decision-makers reading this group. Mine are saying "We've > got to drop Informix like NOW before they force us up sh*t creek and > take away our paddle". In as many words. I fear for my job. I'm > beginning to feel I'd be better off busking. Here are some reasons to stick with IDS: IBM will not force anyone to migrate and has, to be fair, done *exactly* what they said they were going to do. IDS is being developed, maintained and supported, and as far as I can tell, will continue to be. The roadmap has been updated faithfully. Features are being logged for 9.50. IDS tech support is still the best in the business. My main moan is that they gave / give too much credence to analysts who say that if they do anything to promote IDS they will be downgraded because of a confused database strategy. And we all know that the analysts are always right. If there is any agenda, it is a bit of wishful thinking that IDS users will migrate to DB2 en masse because it's just so much better. There are some nudges that keep IDS out of the mainstream, like undercutting the price with DB2 and trying to force everyone onto the wonderful (and cheap!) Pisspoor Disadvantage support scheme. As far as I can tell, IBM likes to have one-size-fits-all processes, and I assume that the reason IDS has to merge into IBM's way of doing things is because of this. I think this is a major recipe for inflexible and intransigent business dealings with them, myself, but I could be wrong. I mean, my dealings with them have been plain sailing... :o# I've not seen any evidence of anything like an active agenda to shaft anybody. Mind you, what do I know. :o) > So, how do we rebuild that confidence? All I hear here is bad-mouthing > and little that's constructive. I am not sure what anyone has said (in this thread anyway) that would cause your management concern. There are still a large number of vocal users who care passionately about Informix and still want it to take over the world. They do not want IBM to force them up a creek and take away the paddle, either. Your management should be rejoicing that this is the case. IBM have not broken any promises (that I can think of off-hand, anyway), all that the moaners are unhappy with is that they did not promise enough. It is their right not to promise enough and it is our right not to accept that lying down. :o) -- "C'est pas parce qu'on n'a rien ''' dire qu'il faut fermer sa gueule" - Coluche

Clown, You make some good points. I was particularly interested in your characterization of how IBM does business, with P****port Advantage, etc. Have you ever done business with Microsoft? It isn't much better, not to mention other big software houses that sell software and yearly maintenance outside of database software, like ETL vendors. So here we are, IBM says attrition of Informix licenses vs DB2 licenses, what's it gonna cost us? I think they said that from the beginning. Just stop the hemmoraging until we can improve DB2 for the non-VM market. Everybody gets that. The other thing that concerns me is your characterization of what constitutes "better technology". As far as SQL-Server and Sybase, they are go-limp-push-the-button databases that do not break very often, if at all, and perform surprisingly in quite a predictable manner. Sybase interestingly enough seems to be coming back from the dead, their CEO steered their ship smoothly through the dot-bomb perfect storm, and now they live another day. I'm not a big fan of either of these database engines, but does it really matter? I get paid equal to my past Informix life, it's just now I work with a Chevy instead of a Rolls Royce. We have yet to log a call to Microsoft about anything with SQL-Server, about a "bug" in the engine. Is this sheer laziness, ignorance, or what? Nope, it's just a well-designed database engine that works well in its space, and has a plethora of features. It just runs and runs and runs. You really don't need a DBA for it, unless you're working with big volumes of data. It won't cluster worht a sh** but that's what other engines are for. It's important to see what each of these engines do, and have for features in order to really say that they are "superior" or "inferior". Admit it or not, MS sells commodity databases, and SQL-Server gets put in places Informix people could only dream about. Honestly, SQL-Server is the right choice because honestly you wouldn't want Informix where a lot of SQL-Server is living. Informix breaks more often, and that's it. Then what? Call IBM? Nope. Install SQL-Server. You don't want to risk your own hide on Informix when you can lower the risk with SQL-Server, it has market acceptance even when it runs like a snail. So is IBM wrong for going after the market the way they are going? The new economy will focus on databases that are simple, robust, and go-limp-push-the-button. If that means less control for the DBA, well, that's the way it goes. Microsoft for all its crappy products sells a lot of product, and for IBM to increase its risk with a database, that for our understanding is "superior" to DB2, makes absolutely no sense to sell in a market that doesn't want that kind of database, nor would it know what to do with it anyway. Most shops today would rather take a performance hit for comfort any day over superior, breakable, Informix. Honestly ain't no way I could abuse an Informix database the way we abuse SQL-Server, just aint no way, unless the latest release has improved, but who cares? People will go with the majority market, not the unknown deathware. Incidentally, think about this, if SQL-Server were ever ported to Linux, most of the database vendors out there would sh*t their pants. SQL-Server on Linux... now THAT would REALLY be interesting. ( Watching the rest of the market sh** their pants on that one would be interesting too! :-) Tim "Obnoxio The Clown" <obnoxio@hotmail.com> wrote in message news:bvtss5$vdo5o$1@ID-64669.news.uni-berlin.de... > Malc P wrote: > > > Malcolm W more or less hit the nail on the head. > > If he did, then why are you asking questions about IBM's IDS strategy? :o) > > > My input to this is, I'm an Informix DBA and I'd rather not have to > > learn how to DBA another vendor's offering. This statement may or may > > not be rational but hey, that's me, and I'm old. > > What I guess we all want to know is whether the impending merge of > > technology (as I recall, the working name quoted at the Infobahn was > > 'DBNew' but what's in a name?) will: > > > > A) retain the capabilities and performance of IDS > > B) be an upgrade which will be no more difficult than a version > > upgrade, i.e. take old server down; install new server; start new > > server and allow conversion to occur; go online. 1.5 hours, max. > > C) Retain the administration capabilities of IDS > > D) run our 4GL/SQL/ESQLC etc without mods. > > > > If any of these aren't met, will the transition at best be no more > > than a weekend's work to iron out? > > Not if the administration doesn't get resolved. Conversion will almost > certainly depend on data set size. I strongly doubt that DB2 will be able > to match the more esoteric aspects of IDS without considerable work, it may > be better to wrap a DB2 wrapper around IDS than try to implement all the > Illustra APIs. > > There is talk that 4GL will be given DRDA (IBM data access) at some > unspecified point in the future, as there is talk that 4GL will somehow be > integrated into IBM's EGL strategy. I'm not aware of any specifics on > either of these. > > Allegations have been made that some of the onstats may be put into a > forthcoming version of DB2, but I'm not sure there will ever be a full > merge of the administration between the two. (That's just an opinion, mind, > I have no concrete reason for saying that.) > > > Further to that; are there any *really* good reasons we shouldn't look > > at Oracle, Sybase, SQLServer etc. if the transition to DBNew is more > > difficult than we'd like. > > We all know that it's plainly obvious that there isn't a really good reason > not to look at these if DBNew doesn't cut the mustard. Price, quality, tech > support, all the usual. > > > And we don't need answers that begin "We all know that...." or "It's > > plainly obvious...." and so on. > > Oh, sorry. > > > Does IBM have a hidden agenda to shaft Informix users? No speculation, > > please. Those of us who are NOT close to the front line, those of us > > who just DBA or write code every day, don't know the wheres, hows or > > whys of the cynicism, but it certainly makes the confidence in our > > chosen technology plummet. > > IBM doesn't have any agenda at all, from my dealings with them. The whole > reason I and several other dishonest and slovenly people are moaning is > because the chosen technology is so much better than anything else and I, > for one, do not feel that IBM is doing IDS justice, for want of a better > term. But to be fair to IBM, this just seems to be the way they work. I've > heard that someone has woken up to the fact that NUMA-Q was actually quite > spiffy technology, and now they are scrambling around trying to find the > Sequent people that they dispersed or lost after the acquisition. Or maybe > they just need them for the fight with SCO. :o) > > > You guys haven't got a CLUE the effect all the bad-mouthing of IBM > > that goes on here is having on IT management teams - there are > > corporate decision-makers reading this group. Mine are saying "We've > > got to dr