3 ms·

> An organization that wins by exercising power starts to lose the ability to win by doing better work. Yup. That's more or less what's happening or Oracle ri

by tharne 5y ago

> An organization that wins by exercising power starts to lose the ability to win by doing better work. Yup. That's more or less what's happening or Oracle right now. They'll profitability soldier on for a while due to the massive amount of legacy servers running Oracle databases at companies all over the world. But, every $BIG_CORP out there has on their roadmap a long-term plan to either drastically reduce their spend on Oracle products, or more often, eliminate it entirely.

beachy 5y ago
> But, every $BIG_CORP out there has on their roadmap a long-term plan to either drastically reduce their spend on Oracle products, or more often, eliminate it entirely. Hyperbole much? I guarantee there are plenty of companies doing giant transaction rates happily and stably on their big Oracle. And plenty of ERP customers who are resigned to staying on their Oracle platform for the foreseeable future, even if only because the competitors are just as greedy shark as Oracle.
- tharne 5y ago
  It really isn't hyperbole. Ask 10 random CTO's about Oracle, and the majority will tell you they're looking for offramps. It's not just a matter of Oracle being greedy, it's that they are also bullying and difficult to work with. And their licensing rules are one of the circles of hell.
  - rconti 5y ago
    And that has been true for 15 years.
    - wizzwizz4 5y ago
      The alternatives have been getting better and better. With free software, you can just use an older version if it's better; with Oracle, the latest version is what gets supported, so it can get worse.
  - cptaj 5y ago
    Saying that and actually putting in the effort to do it are monumentally different things in some orgs
  - jmull 5y ago
    I mean, I'm sure our CTO would tell you they are looking for an offramp. From time to time we start an initiative to do so (i.e., when there is a new set of executives at a certain level). However, as the awareness of the actual scope and cost of migrating away becomes clear, the effort is dropped. It's technically possible, but also risky and no one wants to practically pause other development while it occurs -- which makes sense since that would probably be fatal.
    - virgilp 5y ago
      And you know what tends to happen? Something comes along, that makes it feasible. It might be something surprising. And then, all of the sudden, there's no way to respond - because Oracle the company is so entrenched in the business model of profiting from locked-in customers, that they can't develop a novel response to a novel threat.
gumby 5y ago
> every $BIG_CORP out there has on their roadmap a long-term plan to either drastically reduce their spend on Oracle products, or more often, eliminate it entirely. Oh, I hope you’re right, but am dubious. The F500 (and G500) is wedded to legacy systems, for good reasons, a marriage that even extends to new deployments.
- techdragon 5y ago
  Look, it can be right but sometimes the “long term plan” is on the order of a decade or two. “Let’s replace everything” doesn’t happen in a hurry, so it can take a LONG time to shift. I’ve seen projects that are 5 years into approximately 10 year transitions off MongoDB and MySQL in favour of PostgreSQL because everyone is realistic about the available development time to completely reimplement the core product while continuing to build new product features to maintain customer growth and remain fiscally solvent.
mohaine 5y ago
This has been Oracle's business model forever. The price you pay (off base price that is) is based off off the price they estimate you can pay and what it would cost to migrate off to something else. Also, they purchase 3rd party applications at a regular basic to make sure new customers/hostages are always coming in the front door.
MisterBastahrd 5y ago
You say that, but Oracle hasn't changed in any substantive way in the last 30 years in this regard and they're still insanely profitable. Oracle has the funds to capture key parts of the enterprise ecosystem, so if a new one pops up to replace the Oracle branded product, Oracle will just buy it if Microsoft doesn't first.
yardie 5y ago
> every $BIG_CORP out there has on their roadmap a long-term plan to either drastically reduce their spend on Oracle products, or more often, eliminate it entirely. You can believe this if you want. But $BIGCORP does not have the same license agreements that you and I see. There is a lot of momentum to stay on Oracle and companies have had decades to migrate away. Some have, most haven't and I think not much is going to change in the next decade. [] I'm literally telling a coworker we can't do what she wants to do with a Oracle DB server because the license doesn't allow it. Then I get an email from a VP that states we have a special license that does specifically allow it. I'm left wondering how many other companies have special licenses.
- 300bps 5y ago
  I agree that every large corporation that uses Oracle software has a unique license with Oracle. And every one I’ve heard of considers Oracle a predatory vendor and they want to stop using them. They do things like charging you per CPU to run Java and even though you only run it on four CPUs you theoretically could run it on your entire VMWare cluster so they charge you for 10,000 CPUs. Which is ridiculous so they’ll give you a unique license for their software and they’ll bundle it all together so it isn’t a la carte anymore so even if you stop using 40% of their software you pay the same amount. Oracle’s business model seems to me to be to extract every nickel from their customers until they can figure out how to stop using their software. It’s why there are dozens of open source Java distributions now and it’s why things like MariaDB exist.
  - usefulcat 5y ago
    > And every one I’ve heard of considers Oracle a predatory vendor and they want to stop using them. Ok, but talk is cheap. If you want to know what people actually want, ignore what they say and pay attention to what they do. Oracle's revenue and net income have been pretty steady for at least the past 5 years. Unless they've found some major new revenue stream, it certainly doesn't look like their customers are leaving in droves.
    - 300bps 5y ago
krnlpnc 5y ago
Unfortunately I think this is a naive take, because it's focused on technical superiority. That's only part of the story. Companies partner with massive vendors like Oracle and jump through their hoops to remain in support contracts because it enables them to defer some liability and satisfy various regulatory checklists. Think about a database instability scenario for an online retailer, and consider if you'd rather be the CTO who is to blame for self-hosting, or the CTO who can defer blame onward to the vendor. It's an unfortunate reality in many "enterprise" environments.
- jamesrr39 5y ago
  sure, there are companies that stay on Oracle, and Oracle the company isn't going away any time soon. But other companies that were on Oracle have moved/are moving away to other enterprise-palatable services (e.g. AWS), and new companies rarely pick Oracle. Self-hosting is certainly not the only choice nowadays, there are many other companies offering managed databases/other cloud products.
- LeifCarrotson 5y ago
  The interesting thing is that it's not a financial question of blame for the CTO, it's a reputation problem. It's unlikely that even Oracle would pay to make someone whole after a data loss incident. But the reality is that the ability to blame Oracle or another black-box product from some vendor instead of your in-house instance of some open-source project (that no one outside of engineering has heard of) has real career-altering value to stakeholders. Maybe unjustified, but the effect exists nonetheless.
boldBS00 5y ago
That’s more or less what’s happening to VCs as the attention economy dies, people normalize to shovelware apps being copy-paste attention grabbers of little novel utility. PG is an Apple relative to the majority. His investments create waves in agency of others with or without their informed consent. The antitrust bill Y com signed onto is to help get “web3” on devices. True antitrust would be requiring unlocked OS agnostic hardware. We’ll have to settle for VC backed NFT market places, since that is who will flood App Store agnostic OS’s with options. $BIG_VC has a new metaphor to sell. BOINC2.0!! I mean; web3! They’ll sign up to improve the status quo for them. No need to push for open devices; they’re fine with the OS monopoly.

by tharne 5y ago

beachy 5y ago

> But, every $BIG_CORP out there has on their roadmap a long-term plan to either drastically reduce their spend on Oracle products, or more often, eliminate it entirely. Hyperbole much? I guarantee there are plenty of companies doing giant transaction rates happily and stably on their big Oracle. And plenty of ERP customers who are resigned to staying on their Oracle platform for the foreseeable future, even if only because the competitors are just as greedy shark as Oracle.

tharne 5y ago
It really isn't hyperbole. Ask 10 random CTO's about Oracle, and the majority will tell you they're looking for offramps. It's not just a matter of Oracle being greedy, it's that they are also bullying and difficult to work with. And their licensing rules are one of the circles of hell.
- rconti 5y ago
  And that has been true for 15 years.
  - wizzwizz4 5y ago
    The alternatives have been getting better and better. With free software, you can just use an older version if it's better; with Oracle, the latest version is what gets supported, so it can get worse.
- cptaj 5y ago
  Saying that and actually putting in the effort to do it are monumentally different things in some orgs
- jmull 5y ago
  I mean, I'm sure our CTO would tell you they are looking for an offramp. From time to time we start an initiative to do so (i.e., when there is a new set of executives at a certain level). However, as the awareness of the actual scope and cost of migrating away becomes clear, the effort is dropped. It's technically possible, but also risky and no one wants to practically pause other development while it occurs -- which makes sense since that would probably be fatal.
  - virgilp 5y ago
    And you know what tends to happen? Something comes along, that makes it feasible. It might be something surprising. And then, all of the sudden, there's no way to respond - because Oracle the company is so entrenched in the business model of profiting from locked-in customers, that they can't develop a novel response to a novel threat.

gumby 5y ago

> every $BIG_CORP out there has on their roadmap a long-term plan to either drastically reduce their spend on Oracle products, or more often, eliminate it entirely. Oh, I hope you’re right, but am dubious. The F500 (and G500) is wedded to legacy systems, for good reasons, a marriage that even extends to new deployments.

techdragon 5y ago
Look, it can be right but sometimes the “long term plan” is on the order of a decade or two. “Let’s replace everything” doesn’t happen in a hurry, so it can take a LONG time to shift. I’ve seen projects that are 5 years into approximately 10 year transitions off MongoDB and MySQL in favour of PostgreSQL because everyone is realistic about the available development time to completely reimplement the core product while continuing to build new product features to maintain customer growth and remain fiscally solvent.

mohaine 5y ago

This has been Oracle's business model forever. The price you pay (off base price that is) is based off off the price they estimate you can pay and what it would cost to migrate off to something else. Also, they purchase 3rd party applications at a regular basic to make sure new customers/hostages are always coming in the front door.

MisterBastahrd 5y ago

You say that, but Oracle hasn't changed in any substantive way in the last 30 years in this regard and they're still insanely profitable. Oracle has the funds to capture key parts of the enterprise ecosystem, so if a new one pops up to replace the Oracle branded product, Oracle will just buy it if Microsoft doesn't first.

yardie 5y ago

> every $BIG_CORP out there has on their roadmap a long-term plan to either drastically reduce their spend on Oracle products, or more often, eliminate it entirely. You can believe this if you want. But $BIGCORP does not have the same license agreements that you and I see. There is a lot of momentum to stay on Oracle and companies have had decades to migrate away. Some have, most haven't and I think not much is going to change in the next decade. [] I'm literally telling a coworker we can't do what she wants to do with a Oracle DB server because the license doesn't allow it. Then I get an email from a VP that states we have a special license that does specifically allow it. I'm left wondering how many other companies have special licenses.

300bps 5y ago
I agree that every large corporation that uses Oracle software has a unique license with Oracle. And every one I’ve heard of considers Oracle a predatory vendor and they want to stop using them. They do things like charging you per CPU to run Java and even though you only run it on four CPUs you theoretically could run it on your entire VMWare cluster so they charge you for 10,000 CPUs. Which is ridiculous so they’ll give you a unique license for their software and they’ll bundle it all together so it isn’t a la carte anymore so even if you stop using 40% of their software you pay the same amount. Oracle’s business model seems to me to be to extract every nickel from their customers until they can figure out how to stop using their software. It’s why there are dozens of open source Java distributions now and it’s why things like MariaDB exist.
- usefulcat 5y ago
  > And every one I’ve heard of considers Oracle a predatory vendor and they want to stop using them. Ok, but talk is cheap. If you want to know what people actually want, ignore what they say and pay attention to what they do. Oracle's revenue and net income have been pretty steady for at least the past 5 years. Unless they've found some major new revenue stream, it certainly doesn't look like their customers are leaving in droves.
  - 300bps 5y ago

krnlpnc 5y ago

Unfortunately I think this is a naive take, because it's focused on technical superiority. That's only part of the story. Companies partner with massive vendors like Oracle and jump through their hoops to remain in support contracts because it enables them to defer some liability and satisfy various regulatory checklists. Think about a database instability scenario for an online retailer, and consider if you'd rather be the CTO who is to blame for self-hosting, or the CTO who can defer blame onward to the vendor. It's an unfortunate reality in many "enterprise" environments.

jamesrr39 5y ago
sure, there are companies that stay on Oracle, and Oracle the company isn't going away any time soon. But other companies that were on Oracle have moved/are moving away to other enterprise-palatable services (e.g. AWS), and new companies rarely pick Oracle. Self-hosting is certainly not the only choice nowadays, there are many other companies offering managed databases/other cloud products.
LeifCarrotson 5y ago
The interesting thing is that it's not a financial question of blame for the CTO, it's a reputation problem. It's unlikely that even Oracle would pay to make someone whole after a data loss incident. But the reality is that the ability to blame Oracle or another black-box product from some vendor instead of your in-house instance of some open-source project (that no one outside of engineering has heard of) has real career-altering value to stakeholders. Maybe unjustified, but the effect exists nonetheless.

boldBS00 5y ago

That’s more or less what’s happening to VCs as the attention economy dies, people normalize to shovelware apps being copy-paste attention grabbers of little novel utility. PG is an Apple relative to the majority. His investments create waves in agency of others with or without their informed consent. The antitrust bill Y com signed onto is to help get “web3” on devices. True antitrust would be requiring unlocked OS agnostic hardware. We’ll have to settle for VC backed NFT market places, since that is who will flood App Store agnostic OS’s with options. $BIG_VC has a new metaphor to sell. BOINC2.0!! I mean; web3! They’ll sign up to improve the status quo for them. No need to push for open devices; they’re fine with the OS monopoly.