Slashdot Mirror
U.K. Government Seeking To End Reliance On Oracle
jfruh writes: The U.K. Cabinet Office has reportedly asked government departments and agencies to try to find ways to end their reliance on Oracle software, a move motivated by the truly shocking number of Oracle licenses currently being paid for by the British taxpayer. The Department for Environment, Food and Rural Affairs alone has paid £1.3 million (US$2 million) per year for some 2 million Oracle licenses, or about 200 licenses per staff member.
200 licenses per year? If anything, that doesn't speak to technical concerns. It points towards incompetent legal / licensing / contracting. Who's negotiating those licenses with Oracle? Do they know what they are doing at all?
Just because I can hook a shark from a boat, I do no offer to wrestle it in the water.
$1USD per Oracle licence! How do we do it? Volume!
I'm still walking funny from the last time we negotiated with Oracle.
DEFRA is apparently paying $2M for 2M licenses, which if my math is correct, works out to $1/license. Sounds much less reasonable when you see it as 200 licenses per employee, but not that bad when you think of it as $200/year per employee in licensing fees. If this is completely unreasonable, then save money by firing employees who cost at least 2 orders of magnitude more than $200/annum and are not productive. At a minimum, if there really is no need for 200 licenses per employee, the first heads on the block should be the people who promoted/approved the contracts which got them 200 licenses per employee. Once again, it is still $200/employee each year, so what is the big deal?
Considering they are only paying $1 per license on average each year, framing the problem with a license per employee count is very misleading. The article should have focused on them spending $200 yearly on licenses per staff member. Or under $17 per month per staff member. Doesn't sound nearly as bad in this context, but then again the true point of the article was to get page views. This shows why I'm not in marketing.
-- All that is necessary for the triumph of evil is that good men do nothing. -- Edmund Burke
Conveniently located next to the Ministry of Silly Walks....
love is just extroverted narcissism
That sounds good and all, but they're still going to want to pay some company for support. Any government that's too incompetent to manage licensing properly should not be trusted with supporting their software.
I don't think the government needs to strictly use open source, but I do agree that anything they use should be an open standard as that allows for the most competition in terms of choice of support provider, contractor, etc. as well as ease of transitioning from one to another. Also, if they want to buy custom software, the code must be open sourced (or the government must be given a license to treat it as such from their perspective) so that they're not locked into a single company for ongoing support, bug fixes, or future enhancements.
Oracle products are specifically designed to make it very difficult and costly to leave the platform given all their proprietary extensions to SQL and supported programming language and development tools.
If your application was designed with Oracle development tools you are likely completely S**t Outta Luck. But if all you did was use Oracle as an RDMBS and avoided all their lock-in traps you should be able to port to PostgreSQL.
But in most situations, Oracle is the Hotel California of platforms: "you can check in anytime you want, but you can never leave.." at least not without significant costs in porting which will be more painful and risky than to simply keep paying.....
Because of this the best option is usually to specify and enforce that Oracle *NOT* be used on any new or replacement projects while the organization just keeps paying and paying and paying on the systems that require Oracle.
There are a number of very good reasons that few Internet startups run out and buy Oracle for infrastructure use.
Facebook is billions of individual "Skinner Boxes." And if you use it you are the pigeon!
They should replace it with a nice free, open source solution like MySQL Enterprise Edition to get paid support. Then they'll never have to pay Oracle another penny (Or pence or whatever they call it in the UK)
Have you ever dealt with Oracle licensing, or are you just spouting off?
Because I can guarantee you, Oracle fucks over corporations and governments about the same.
Oracle is legendary for this stuff.
I can also tell you for a lot of applications, the choice is Oracle, and SQL server. And no matter what Microsoft tells you (who is also trying to fuck you over on licensing), for many applications SQL server just can't do the same job.
The vendor of the software needing the DB won't support your open source platform, or anything else.
If you haven't heard about Oracle's licensing practices or think this is inept governments, then you really have no idea of what you're talking about.
Lost at C:>. Found at C.
You do realize that Oracle owns MySQL, right?
Facebook is billions of individual "Skinner Boxes." And if you use it you are the pigeon!
I'm fairly certain that, if a major government like the UK were to go to PostgreSQL, the maintainers of the project would soon find themselves in court over some nebulous IP claim about how some obscure library uses the same call declarations as Oracle's code.
We're in the age where the big software companies have essentially become robber barons.
The world's burning. Moped Jesus spotted on I50. Details at 11.
I manage a team of database developers and database administrators. We live Big Data. We breathe Big Data. Big Data is everything to us. If the data isn't Big, we don't touch it.
We only use the best tools of the trade, and those are NoSQL tools. I know some people like to joke about NoSQL being "web scale", but it's no joke. In our experience, NoSQL is the only way to really work with Big Data.
A good rule of thumb is that if you're using SQL, you're working with Small Data. I was at a conference last year, and some schmuck started talking to me about his 2 peterbyte database. He said his team used Postgrass and SQL. It doesn't matter how big your database is! If you're using SQL then you aren't working with Big Data! 2 peterbytes of SQL data is way smaller than 2 peterbytes of NoSQL Big Data.
It makes no sense to me why anyone would use SQL databases. They are old tech. They aren't the latest and greatest, like NoSQL databases are. Like the CAP Theorem states, NoSQL databases are better because they're "Capable of handling Big Data", "Always the best choice for Big Data", and "Perfect for Big Data".
It's 2015 now. We have better tools available to us than we had in 1975. You don't need to use SQL databases any more. Use a NoSQL database, and get all of the benefits it gives you, including the CAP Theorem. Big Data is important, so you should only trust it to NoSQL databases.
You would be amazed at the number of people who rip off int main(int argc, char **argv) and present it like its their own work. There ought to be a law. Don't even get me started about the whole stdio business.
I've had the fortune (or misfortune depending on your definiton) to work on a lot of companies' systems and have had a very "cross platform" career. Oracle's licesning, which has gotten worse in recent years, is just now starting to send most companies looking for other ways to do the same thing. The problem is that Oracle is still the de facto standard for "enterprisey" software projects. A lot of this is legacy -- for quite some time the only mainstream database systems were DB2 on AIX or pSeries/zSeries, and Oracle on Solaris. You might say that's ancient history and you're right -- SQL Server is good enough for most workloads that need a "fully supported" DB and Linux is a viable alternative to Solaris. But I can tell you that these applications don't just die -- they're alive and more functionality is being built on top of them. Most big enterprise applications (SAP, PeopleSoft, JD Edwards, and so on) are either Oracle products or are integrated to run on Oracle middleware/databases. Most of the big outsourcing firms' "standard stacks" revolve around Oracle DB running on Linux or Solaris, and J2EE running WebLogic. This makes perfect sense; outsourcers can pick up CS grads who know Java for cheap, and J2EE's nature lets you parcel out and offshore pieces to whoever is cheapest that week.
Since most government IT is outsourced both in the UK and the US, I would say that it would be very difficult to replace Oracle without re-architecting whole applications. Some stuff is easy - you don't need a Solaris license to run Apache for example. Some is not -- just like SQL Server, Oracle makes it very easy to slip into "Oracle-only" development mode when interacting with databases and middleware. Once that dependency is in place, it either has to be identified and pulled out, or it just keeps chugging along. And since systems like this are not sexy (customs processing, DMV records, tax collection, etc.) they don't get seen by the public very much.
Postgres was started from The guy that received and ACM Turing awards for his works database :
Michael Stonebraker has made fundamental contributions to database systems, which are one of the critical applications of computers today and contain much of the world's important data. He is the inventor of many concepts that were crucial to making databases a reality and that are used in almost all modern database systems. His work on Ingres introduced the notion of query modification, used for integrity constraints and views. His later work on Postgres introduced the object-relational model, effectively merging databases with abstract data types while keeping the database separate from the programming language. Stonebraker's implementations of Ingres and Postgres demonstrated how to engineer database systems that support these concepts; he released these systems as open software, which allowed their widespread adoption and their code bases have been incorporated into many modern database systems. Since the pathbreaking work on Ingres and Postgres, Stonebraker has continued to be a thought leader in the database community and has had a number of other influential ideas including implementation techniques for column stores and scientific databases and for supporting on-line transaction processing and stream processing.
You do realize that Oracle owns MySQL, right?
They should run it on Sun hardware to stay even farther away from Oracle.
Pretty soon they'll just be able to move all their databases and schemas and stuff to systemd. Problem solved!
Just cruising through this digital world at 33 1/3 rpm...
yeah you can even save more by putting it on VirtualBox too.
Harrison's Postulate - "For every action there is an equal and opposite criticism"
I wont say which one it was, because the walls have ears, but I worked at an Australian govt department and we where doing just that, moving what we could over to Postgres
The big problem was Financials. There just isn't a replacement that'll suffice at a government level, so theres still a bit of stickyness in that area.
Mostly though we where doing a lot of our stuff in modern MVC stuff and phasing out a lot of crufty java and oracle stuff, and thats a pretty good time to start reducing the oracle crackpipe addiction
Excuse the Unicode crap in my posts. That's an apostrophe, and slashdot is busted.
EnterpriseDB nearly costs as much per server as Oracle does.
I'm confused by the units here. WTF is a dollar squared?
A dollar. /ducks
The Future of Human Evolution: Autonomy
What you say was true for pgsql 8, in 9 you can have hotstandby www.postgresql.org/docs/9.3/static/hot-standby.html. Also some application delivery controller can analyze sql statement on the wire to enable replication.