Re: Informix running on Linux

Nils Myklebust wrote: > snip > Well, nothing is free - not even Linux, but it's cheap at least. > I paid nothing directly for my copy. However I guess you could argue download time, ISP cost, hardware, etc. But the software was downloaded and installed for absolutely FREE. snip > :You may recall a few years back SCO was bundling Ingres with one of the > :first System V releases when they were breaking away from Xenix. It > :flopped because nobody cared about Ingres. Interesting that Informix > :has ultimately bought Ingres in Dr. Stonebreaker. > > Well, I still think that Informix buying Illustra was a very smart > move. If they hadn't had it their long term prospects would have been > worse - much worse. Now their long term prospects are very good > indeed, particularly if they also regain their previous understanding > of the low end market. > I didn't mean to imply that this was a bad thing. Ingres is back, in the form of Postgres->Illustra->Universal Server. I see it as an interesting twist of fate. > ...snip... > > :Linux on PC --> make mo money --> upgrade to Linux on Digital Alpha, > :retain investment in software, and OS. > > I think you have overestimated the Alpha machines. I doubt very much > that the performance will be much different except on the largest > Alphas. The Intel Pentium processors are becoming fast, and the disk > technology is essentially the same. (Check the TPC/C results, they > also show this - for what they are worth.) > Could be... But I think the point was it's an upgrade path that's unique. Most of the people I would target for Informix on Linux are businesses and organizations who cannot afford the total cost of ownership of these other platforms. Informix can get name recognition out of Linux without any advertising--the word of mouth announcement alone would be enough. The idea is interesting that you could start out with a Linux system on PCs, then migrate to "bigger" boxes, such as DEC Alpha. > > I agree on the tools bussiness. Much of it will probably even be > portable to other Unixs. > You are however wrong on Unixware. SCO bought that from Novell, and I had no idea... Novell has their own pain right now, trying desperately to hang on. It's interesting that so many big companies took a dive in recent months and years, missing the market so badly. > > May be it's just me feeling better with an SE installation at remote > sites with no computer literate users? Soon I will try the OWS for > this purpose. > NT is so expensive though... be prepared to open your wallet. > ...snip... > Obviously they should do a real port. My point was only that some > version would be better than non at all. > I understand. > ...snip (developing for MS Windows using a Linux server)... > > :And another valid point: Most of the success we've seen with Informix > :has been on UNIX. Client/server success seems to follow a UNIX server > :implementation of Informix, and client PCs hitting it across the pipe. > > The above is true. Your following items on NT isn't neccessary to show > this. Informix on NT is too new yet. I think it can be very > successfull, but I also think they need a realy low priced version of > Universal Server with some high quality development support to do > that. That is however an entirely different issue. > NT and Windows 95 were both originally introduced as the panacea UNIX killers. We all went through the UNIX-is-dead era, thinking NT was going to replace UNIX. This is the whole point to Linux. NT has failed miserably as a cost-effective solution. Most serious NT 4.0 Server installations come with at least two processors, and 128 MB RAM. The disks better be SCSI, and even then you won't know how well it'll run until you test your software on it. Yes, I poopoo NT because it has failed to deliver the multi-tasking features UNIX enjoys at a cost-effective price on x86 platforms. Many sites now require an x86 system that is so high-end just to run NT it has gone way beyond it's dream of a cheap multi-tasking Windows platform. NT is a money pit. It remains relavent to this discussion because of Linux. If NT were so great, we wouldn't even be considering Linux. The argument for Linux would be moot. NT was originally promised to be an inexpensive multi-tasking server. Small businesses were anticipating a real sweet operating system that would replace UNIX. It didn't happen. MS is interested in nothing but upgrades and corporate dollars. Even if Informix stays with NT, it could be short lived if the migration to NT 5.0 dies. snip > : > :This is dependent on NT surviving the hit the Linux port will make on > :NT. NT stands to lose big ground in the next few months as people > :find out more and more how much a security risk the machine is. And > :Active X with permeat the NT 5.0 release, which is something I will > :not use. It'll just perpetuate the Microsoft grip. I'd rather see > :the emphasis be that Informix scales the product as much as they > :scale the user count on the engine. Linux makes this possible. Instead > :of using NT as the metric, use Linux. It works, and is open. > > You are of course wrong on NT. This will not make a significant impact > on it. NT is simply to much in demand. Both will grow. It wouldn't be the first time I'm wrong about NT, but the future of NT is full of more uncertainty from MS. This I cannot be wrong on. I've spent way too much money building an NT system so I can develop and learn web applications. But for what it costs, and what it does, it's too expensive. It'll make a killer Linux machine. > Informix have to keep their momentum in the NT market as well, and do > *nothing* to even hint at any negative view of it. Try to see things > from Informix's side a little, and you will easily see why. > I'm quick to like NT for many reasons. However, I also know that if I want to do anything on it, I have to basically work ON it, and not remotely. It is a real client/server product. But it has much that is either documented poorly or not at all, and no other vendor sells it or supports it. We're always hearing how Linux is unsupported, yet I can find more resources for Linux in seconds compared to NT, or Solaris. > It's *extremely* dangerous to make this an Linux versus NT or anything > else issue. If Informix starts to believe they will loose sales on NT, > why should they port to Linux. Success on NT is strategically > important to them. That's not the case with Linux. Not yet anyway. I poopoo NewEra all the time, NT doesn't get any special treatment from me either. If Informix is depending heavily on NT sales, then you should sell your stock now before it hits $1.00. NT is a pooch. It wouldn't exist if a true low-cost UNIX had the same availability of software that windows has enjoyed. Linux is coming up though. NT should not be viewed as superior to Linux. This is the whole point. I agree it's not Linux vs NT, however NT was touted as a small business machine. Microsoft does not understand what commercial UNIX is all about