Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Oracle vs. PostgreSQL – A Comment

postgresql.org

Post reply on HN
Print view View original

211–220 of 238 posts

Re: Oracle vs. PostgreSQL – A Comment

#211

Post by gurjeet »

jdm2212 wrote: post #46
paulryanrogers wrote: post #42

For those still on Oracle, why? Is it regulations or executives demanding certified solutions? And if so then with so many bugs at what point does certification become a negative signal?

Query hints are one reason. I'm not a fan of Oracle as a company, but for certain usage patterns you need query hints and Postgres doesn't support them.

EnterpriseDB's Advanced Server (based on Postgres) has the query optimizer hints feature [1].

Disclosure: I implemented the Optimizer Query Hints feature at EnterpriseDB.

[1]: https://www.enterprisedb.com/edb-docs/d/edb-postgres-advance...

Re: Oracle vs. PostgreSQL – A Comment

#212

Post by dwheeler »

GordonS wrote: post #22
phoe-krk wrote: post #3

Copying from https://stackoverflow.com/questions/12115397/is-it-against-l... : The standard license ( http://www.oracle.com/technetwork/licenses/standard-license-... ) you agree to when you download software from the Oracle Technology Network (OTN) does state that you're not allowed to disclose benchmarks. You may not: > - disclose results of any program benchmark tests without our prior consent. I sadly suspect that…

> You may not: > I mean... does anyone here think this is in any way reasonable? Is this even legal?

I think this clause is not reasonable and should not be legal. However, these clauses (they're called "DeWitt clauses") have been around for almost 40 years and they have not gone away.

For more, see my essay, "The DeWitt clause’s censorship should be illegal" https://dwheeler.com/essays/dewitt-clause.html (I mentioned it in another thread, but you might not have seen it).

Re: Oracle vs. PostgreSQL – A Comment

#213

Post by dwheeler »

ethbro wrote: post #62
GordonS wrote: post #27

Earlier quoted context omitted.

I can't argue with that. I suppose regardless of how ridiculous it is, Oracle has very deep pockets and is infamously litigious, and the US legal system doesn't help - given that, I doubt legality really matters...

You can't sue someone for violating ToS... right? At worst, you can refuse to do business with them and revoke any current contracts. You'd have to sue under actual law, which would presumably be libel, which is (in)famously hard to prove in the US [1]. If he had a good faith belief that what he reported was true, that case goes nowhere. Indeed, in lieu of a signed and dated note declaring his knowledge that his benc…

A company absolutely can sue someone for violating the terms of service. For huge amounts of money, too, and it might be both you and your organization who have to pay. In the US it would probably be at least considered a contract violation & a copyright violation, not just libel. Contract law has a lot of teeth, and copyright violations have massive statutory damages. Oracle has a massive legal department; they can probably find a thousand other laws it potentially violates without blinking an eye.

I really don't like DeWitt clauses, I think they're horrific. But there's a reason people generally avoid fighting them in court - it's very expensive with an uncertain outcome.

Disclaimer: I'm not a lawyer.

Re: Oracle vs. PostgreSQL – A Comment

#214

Post by gurjeet »

jacques_chester wrote: post #196
phonon wrote: post #195

Earlier quoted context omitted.

Check out https://github.com/arkhipov/temporal_tables or https://github.com/scalegenius/pg_bitemporal

I'm familiar with these and they are useful, but I would greatly prefer mainline support using the SQL standard. A mainline implementation is far more likely to lead to optimisations being made to the query planner, indices, disk formats etc.

You may be pleased to know that someone is actively working on this. There was a presentation [1] by the author Paul A. Jungwirth at this year's PGCon last week.

[1]: https://www.pgcon.org/events/pgcon_2020/schedule/session/100...

Re: Oracle vs. PostgreSQL – A Comment

#215

Post by perl4ever »

jrockway wrote: post #134
bmn__ wrote: post #96

Earlier quoted context omitted.

> It would be inconvenient, but not an existential crisis. Don't generalise from your life; you only have to take care of yourself and Ayaya. A majority of people are deeply committed to a relationship/family/tribe, where losing all one's money definitely constitutes a crisis.

Hey now, leave my Ayaya out of this. Here is my point -- the potential losses are bounded. They can't unteach you what you've learned in 30 years of work, so you can probably earn income even after losing the biggest lawsuit EvilCorp can muster. Or go work at Starbucks if you are somehow banned from using computers, which is not really something that can happen for violating a software EULA. Compare this to driving t…

Every now and then you read a news story about something that you might expect would be a civil litigation matter becoming a criminal case.

If you are going around thinking "potential losses are bounded" you should read some of those stories and think about tail risk.

Corporations seem benign when you take for granted that their power isn't directed by anyone who has it in for you personally, and their employees are just going through the motions.

Re: Oracle vs. PostgreSQL – A Comment

#216

Post by dang »

quietbritishjim wrote: post #110
jrockway wrote: post #82

Earlier quoted context omitted.

People are scared of corporations, but they don't really have much ability to damage you. They can't cause physical harm to your body. They can't take away your loved ones. All they can do is take some money. You started life with $0, so you can probably make it all back. It would be inconvenient, but not an existential crisis. The likeliest outcome is that nothing will happen. The second most likely outcome is that…

Your first paragraph is totally ridiculous. Maybe you started with $0, but maybe that was 30 years ago and losing all your savings from then till now, including your home, is more than an inconvenience. If you have dependants then that loss could directly impact their lives and especially their education – what if some of that money had been for a college fund? In the US especially, and also to some extent in Europe,…

When disagreeing, please reply to the argument instead of calling names. "That is idiotic; 1 + 1 is 2, not 3" can be shortened to "1 + 1 is 2, not 3."

https://news.ycombinator.com/newsguidelines.html

Your comment would be fine without the first sentence.

Re: Oracle vs. PostgreSQL – A Comment

#217

Post by jpalomaki »

phoe-krk wrote: post #3

Copying from https://stackoverflow.com/questions/12115397/is-it-against-l... : The standard license ( http://www.oracle.com/technetwork/licenses/standard-license-... ) you agree to when you download software from the Oracle Technology Network (OTN) does state that you're not allowed to disclose benchmarks. You may not: > - disclose results of any program benchmark tests without our prior consent. I sadly suspect that…

Maybe one justified reason is that proper benchmarking can be hard. Especially running fair benchmarks between competing complex products.

At some point some big database vendors were showing full-page ads in prominent industry publications comparing database X to Y. This kind of clause in license may make publishing misleading performance figures a bit harder.

Re: Oracle vs. PostgreSQL – A Comment

#218

Post by F_J_H »

paulryanrogers wrote: post #42

For those still on Oracle, why? Is it regulations or executives demanding certified solutions? And if so then with so many bugs at what point does certification become a negative signal?

Oracle Application Express (APEX).

Comes as a no-cost option with the database, making Oracle the only complete, end-to-end, data management system I know of, where you can build and deploy rich web apps, "out-of-the-box", without installing or integrating anything else.

A friend of mine (experienced developer) built a web app with React, (CRUD and Dashboard) and it took him the better part of a day. Built the same thing with Apex in about 15 minutes.

That's a big reason we use it. Simply can't stand-up data centric, business focused web apps with as quickly with anything else.

YRMV

Re: Oracle vs. PostgreSQL – A Comment

#219

Post by F_J_H »

Meh.

We use Oracle's (newish) autonomous database offering, and haven't really experienced any bugs at all. No worries about backups, very little DBA time, and the on-demand licensing makes it quite economical. We just get on with it, and couldn't be happier.

Of course, YRMV.

Re: Oracle vs. PostgreSQL – A Comment

#220

Post by redis_mlc »

1) FYI: Oracle Enterprise is basically dead in Silicon Valley for new projects (because it doesn't scale financially.) Every Oracle DBA I know transitioned to MySQL over a decade ago, and they look at you funny when you ask about Oracle.

2) Oracle Enterprise is free to use for dev/qa, but each prod instance costs money.

3) The contraint on publishing benchmark results is not what devs think. It's to prevent poorly conducted bm efforts, not really to hide anything. A vendor can't run around defending itself against non-DBAs doing benchmarks. Most of the links to results on HN admit, "We tuned our fave database, and used the defaults for the other database." Seriously.

Having said that, you can publish benchmark scripts, which is what MySQL did. That script moved the whole industry forward wrt reliability and performance, as MySQL was the only database that could pass initially. :)

Source: DBA.

Post reply on HN