Why the sun shines for Oracle and it’s Cloudy for others
vanpupi.stepi.net
vanpupi.stepi.net
To connect from node on a mac unlike postgres, mysql or sqlserver where you just install a module (that may or may not need to compile via node-gyp) you must
1. make an oracle account 1a. make sure you prassword donesn't contain <>();%or \ 1b. if your password is too long then oracle truncates them either on storage or verification but not both 1c. try get your co worker to try to log into his oracle account on your computer 1d. just use his account on his computer 2. download the various versions of instant client you need 3. follow these obtuse [1] install steps 4. add a tnsnames.ora file somewhere 5. when login fails update your host file to match the name of current host system
To say I would never use this voluntarily is an understatement, the only reason we are using this is because somebody already on oracle is handing us a wad of cash to use oracle, I don't really see those people who use it being the types who need a free trial to get onto the cloud, people use oracle because they can't quite oracle.
1. https://github.com/oracle/node-oracledb/blob/master/INSTALL....
Oracle tools are not optimized for developers of today. A key selling point to enterprises is your developers could be more productive with their point and click IDE/interfaces.
These days with the tooling around open source I believe you can get things done much faster (and have done this) the code is more maintainable etc.
Second disadvantage consider oracles commerce framework atg. They have their own concepts and terms for things that are in open source solutions making it confusing for developers familiar with open source frameworks. This could be on purpose to sell more consulting for partners.
Next the is the cost of training and up skilling junior devs. With open source you could probably teach yourself try searching for oracle atg tutorials. Check the # of stackover flow questions vs magneto.
Fourth: I have seen people who are "oracle developers" (fusion, odi, erp, atg etc) are mostly just product specialists. Give them a text editor and tell them to write something and deploy to a paas , commit to version control and they struggle.
Finally the software tends to be ops heavy encouraging bad practices like no version control or automation.
All of these factors are not considered in the purchase or "ROI" calculation, but some of them could seriously hurt your return.
I have seen companies spend millions in implementation of an oracle product only to use < 10 % of features and the same business objectives could have been achieved at the fraction of the cost.
* excuse typos on mobile
Running Oracle databases on Windows doesn't just suck, it's extremely more expensive for licensing. Oracle almost forces you to instead set up Oracle Linux servers for this purpose.
I've also been seeing Microsoft reaching out to Oracle customers trying to incentivize them to switch to Azure/Windows/SQL Server. "We'll throw in some money for the engineering effort to switch!", things like that.
"So for Oracle in my opinion, the sun is still shining on premises and I do hope for them the clouds will come, but the way it is now, I’m afraid they‘ll miss this train."
It's way beyond that. Larry Ellison didn't just miss the train, he said where the train was going was crap, and sat on sidelines for way too long. They had a chance early on but LE thought he was smarter than everyone else.
Oracle's saving grace is it's legacy business. Lots of older and bigger companies have Oracle and they can upsell those customers.
But every day a new tech company is born and none of them are adding Oracle. Baring a minor miracle it will be a slow but still profitable decline. There isn't much of a chance they will catch up, much less keep pace.
http://www.vesterli.com/2017/02/14/losing-hearts-and-minds/
key quote: "If you are an Oracle developer with more than 10 years until retirement, I advise you to start planning for your time after Oracle."
It's too bad, I really like Oracle APEX, (the rapid web and mobile development framework that is included with the DB) since you can build and deploy highly functional/rich apps very quickly, which is the big thing end users care about. (i.e. the same reason foxpro and MS Access were so popular).
Anecdotally, I've been hearing from my consulting peers that companies are actively looking to move off Oracle as they don't like being "bullied onto the cloud", and are starting to look at the migration tools in EDB Postgres. That might be a good source of consulting revenues going forward...
That said, dealing with Oracle the company is painful, and I wouldn't do it if I were by myself or running a small shop. Fortunately, I work for BigCo, and we have People who do that sort of thing. By the time it gets to me all the business stuff is worked out, the database is installed, and I'm getting an email with my account information.
So, they have other sources of revenue as well, and can likely build a large revenue stream of subscription services for business applications. (i.e. running PeopleSoft, EBS, etc. in the cloud).